THE SQL Server Blog Spot on the Web

Welcome to SQLblog.com - The SQL Server blog spot on the web Sign in | |
in Search

Joe Chang

  • Server Sizing Comments: the case for 1-socket in OLTP

    Long ago, almost 20 years, there was legitimate concern on whether servers had sufficient capability to drive transaction processing requirements for large businesses. Then for a long period, perhaps from the late 1990’s to 2012, the better choice for most mid to large businesses was to simply buy a 4-way (socket) system. Larger systems were based on a non-uniform memory access (NUMA) architecture that had unusual characteristics and vendors provided almost no information on how to work around issues that could cripple such systems.

    Towards the end of the second era of server sizing, processors (at the socket level) became ridiculously powerful such that properly tuned systems ran at very low overall CPU utilization. However, 2-socket systems in this period did not have adequate memory or IO capability, unless it was a Xeon 7000 or later E7 series system with 2 sockets populated. And even then, the structure of SQL Server per processor licensing meant that many people opted for the 4-way system regardless of requirements.

    In 2012, Intel finally put powerful processors (Sandy Bridge, up to 8-cores) with larger memory and IO capability into the Xeon E5 line (there were both 2 and 4 socket versions). This coincided with SQL Server 2012 per core licensing. Now there was a stronger technical justification and financial incentive to not automatically opt for a 4-socket system with the most number of cores.

    It would seem reasonable to make some effort to first determine the total number of processor cores that will meet requirements with some headroom. Then determine the number processor sockets as more processor sockets means more memory sockets (also considering the difference in DIMM sockets per processor between Intel Xeon E5 and E7 processors). It is unclear whether anyone actually engaged in the technical analysis versus just making the decision strictly on budget goals.

    Today, it is time to consider the astonishing next step, that a single socket system is the best choice for a transaction processing systems. First, with proper database architecture and tuning, 12 or so physical cores should be more than sufficient for a very large majority of requirements.

    We should also factor in that the second generation hyper-threading (two logical processors per physical core) from Nehalem on has almost linear scaling in transactional workloads (heavy on index seeks involving few rows). This is very different from the first generation HT in Willamette and Northwood which was problematic, and the improved first generation in Prescott which was somewhat better in the positive aspects, and had fewer negatives.

    A single socket Intel Xeon E5 processor supports 12 DIMM sockets for 384GB with 32GB DIMMs, or 768GB with 64GB DIMMs which is now at a moderate (50%) premium per GB over the 32GB DIMM. In the old days, DBA’s were taught that memory was precious in reducing disk IO to a tolerable level. That was back when a total system memory of 1GB was considered very large.

    I noticed that after common server memory configuration reached 32GB memory, disk IO could be brought to a very manageable level, well under the IOPS possible with 100-200 15K HDDs.

    Today, all-flash storage is practical. Definitely in direct-attach storage, and maybe in SAN based storage depending on the degree of price gouging from your SAN vendor. So while it might be possible to show that TB-plus memory reduces IOPS from 10,000 to nil, the reality is a modern storage system correctly configured for database applications can easily support 100,000 IOPS if not 1M (which would require NVMe to be practical). In this case, we should find that memory requirements are far lower than the memory capacity of recent generation 2 and 4-socket systems.

    Why not continue to spectacularly over-configured the server system when even with SQL Server per core licensing is only a small portion of the costs in deploying a critical line-of-business application? The big bosses may not question the costs of a 2-socket system or perhaps 4-socket system even factoring in that there may be several environments between production, DR, QA and development. But if performance is inadequate, it may cost you your job.

    The answer is performance. Today, transaction processing performance is heavily dependent on serialized round-trip memory access latency. The term is pointer chasing in that a CPU instruction accesses memory to determine the next memory location to access. No amount of L2/L3 cache can hide this, nor is it possible to pre-fetch memory. (What Intel needs to do is increase Hyper-threading beyond two logical processors per physical core. IBM Power is 4 and SPARC is 8?).

    For a processor with integrated memory controllers (AMD from Opteron on, Intel from Nehalem on), local node memory access round-trip latency is probably 50ns. For memory on an adjacent processor, a 1-hop remote node, it is perhaps 100ns. This information is absolutely crucial in understanding system level performance for transactional databases, and yet vendors try to not mention it, instead stressing metrics that are mostly irrelevant.

    In a single socket system, 100% of memory accesses are local node, because there is only 1 node. In a two socket system, the expectation is that memory accesses are split 50-50 between local and remote node. In principle, it is possible to devise some strategy such that a higher percentage is local, but SQL Server does not make it practical to implement such a scheme. In the Intel Xeon E7 4-socket systems, 1 out of 4 memory nodes is local, and there are 3 memory nodes 1-hop away. In an 8-socket system, 4 nodes are 2-hop away.

    An argument in favor of multi-socket systems is that the number of memory channels scales with the number of sockets. However, in transaction processing, it is memory round-trip latency the dominates, not memory bandwidth. (it is a factor that more memory channels means there can be more memory transactions, but it is unclear as to whether this offsets the increase in memory latency.)

    If someone has reasonably comparable systems with 1, 2 and 4 processor sockets, all of the same processor generation, the impact of the number of sockets can be tested.

    I am also of the opinion that SQL Server cannot make meaningful use of all the processor cores in a top-of-the-line 2-socket (36-cores for Xeon E5 v3), let alone a 4-socket (72-cores) due to most write operations data and log being single threaded. But I would like here opinions on this. Is this addressed in SQL Server 2016 to a degree?

    Addendum
    The above applies mostly to transaction processing applications.
    It is reasonable to suppose that decision support system application could use both the greater number of processors cores and the greater number of memory channels in a multi-socket systems. However it is unclear as to whether SQL Server is effective in implementing parallel execution plans when the degree of parallelism is greater than the number of cores in one socket.

    I would think that the strategy of partitioning and alignment of compute to local (system) node memory from (formerly) PDW applied to a single SQL Server instance, treating each processor socket as though it were a node, which it is, could dramatically improved very DOP parallel execution plans. But it does not seem that Microsoft is interested in this approach.

    I am presuming that column store is better capable of utilizing multi-socket, as it is definitely not encumbered by serialized round-trip memory latency.

    Addendum 2
    The Dell PowerEdge R730 can be purchased with 1 or 2 of 2 sockets populated. The R930 must have 2 or 4 of 4 sockets populated. It appears that the premium from 32GB to 64GB memory modules might be because the 32 is RDIMM and the 64GB is LRDIMM. The 2400MT/s 32GB RDIMM is $558, and the 64GB LRIMM is $1483. For this differential, I would stay single socket, opting for the more expensive 64GB (if necessary) over 2 sockets using the less expensive 32GB, even though I am giving up the 4 extra memory channels.

    If the plan is to implement memory-optimized tables in 2016 (which supports more than 256GB? and has fewer other restrictions) and more than 768GB system memory is needed, then go ahead with 2-sockets. But I would try to get my transaction processing DB under the single socket limit. In the old days, we moved the old out to the DW system.

  • Join and Query hints – serious DBAs need to learn this

    For a long time, I have refrained from endorsing the use of Join and Query Hints. The reason is that there is no such thing as query hints in SQL Server except for a few simple items. The key options listed under Microsoft SQL Server Hints (Transact-SQL) that affect join order, join type, and index usage are not really hints.

    They are in fact directives such that the specified option is employed if possible. But join order optimization is essentially disabled, and the execution plan joins tables in the order as written, exception being EXISTS and NOT EXIST clauses. It is if the Query Optimizer is hyper-sensitive and its feeling are hurt when someone impudently suggests a hint. So if hint is specified, good luck to you. You are on your own.

    Microsoft documentation is obviously very thorough on this matter with the following:
    “Because the SQL Server query optimizer typically selects the best execution plan for a query, we recommend only using hints as a last resort for experienced developers and database administrators.”

    But the fact is that there are complicated queries such that it is difficult to estimate the number of rows, especially past the source table into the intermediate steps. The query optimizer does (edit: not) produce a good plan, because the row estimate is grossly erroneous. This is aggravated by the fact plan costs employed by the query optimizer reflect a fixed IOPS model based around four 1995 era 7.2K hard disks, with no consideration for whether the actual capability of the storage system or if data is entirely in memory.

    So there are times when it is necessary to override the query optimizer. I will not go into the details of query optimization here. Back around 2001, I reversed engineered the execution plan cost formulas for many the fundamental operations without bothering to explain the operations or other aspects of query optimization. Since then, Paul White, Benjamin Nevarez, and others have written very detailed papers on query optimizations. Roughly the strategy is a balance of 1) tables access efficiency, 2) join efficiency and 3) reduce rows as quickly as possible.

    Because join order optimization is disabled when a join or query hint is specified, the key is to write the query to produce a specific join order. There is a method of doing so. See Inside Microsoft SQL Server 2008 T-SQL Querying: T-SQL Querying By Itzik Ben-Gan, Lubor Kollar, Dejan Sarka, Steve Kass on the topic of bushy plans. Once the correct join sequence is found, most of the join types might already be good. Perhaps a few join hints might be necessary. Join and Query hints do no cause the query optimizer to disable index optimization, so index hints should not be needed as frequently.

    This is a much deeper topic than I can discuss at this point in time. So why I am bring it up? Today there was a news report of the cause of the Air Asia flight 8501 crash last December. For details, see CNN Air Asia crash report or Wikipedia

    In brief, there was an equipment malfunction that put the airplane outside of the normal flight envelope. It should have been possible for the flight crew to recover, but they did not do so.

    It further turns out: “because the manual provided by the plane's manufacturer said the aircraft, an Airbus 320, was designed to prevent it from becoming upset and therefore upset recovery training was unnecessary.”

    In our world, we are told the query optimizer “typically” is best, hence we should not have to do manual query optimization, hence we do not need to learn this. So when we do need to do manual query tuning, we do not know how because we did not learn this.

    This is BULLSHIT! I don’t give a damn how good the query optimizer is “typically.” A serious principal DBA needs to learn how to do query optimization so that he/she can do so when the situation calls for it. The starting point is the bushy join style in the Itzik Ben-Gan book.

    Addendum
    here is the link to my previous post on bushy joins, notice at the time I put caveats on it. Bushy Joins
    I am not saying go for broke on join/query hints, just that we need to learn to shape the join sequence. Rob Farley had a great note in the comments. I had seen that query style before, in TPC benchmarks? without understanding at the time what the writer was smoking.

    per Claus comment linking to Dynamic Search Conditions in T‑SQL, I especially like the strategy below, with hard code when good parameters are specified, then some thing else to handle the unusual, be it OPTION(RECOMPILE) or dynamically built parameterized SQL

    IF @orderid IS NOT NULL
    BEGIN
    SELECT ...
    WHERE O.OrderID = @orderid
    AND -- Conditions on Order Details here.
    -- No OPTION (RECOMPILE) here!
    END
    ELSE
    BEGIN
    SELECT ...
    WHERE -- same conditions as before
    OPTION (RECOMPILE)
    END

  • SSD Form Factor and interface

    There is a curious quiet from the enterprise storage community on form factor and interface direction for solid state storage, be it NAND Flash, Intel 3D XPoint or other. On the client-side, personal computing, both desktop and mobile, show clear direction in favoring both the M.2 form factor and PCI-E as the preferred interface for SSD storage. There is a backward compatibility option in M.2 to work with either SATA or PCI-E interface, but I do not think this will be widely used. SATA or hard disks will not go away, only the primary SSD is M.2 form factor, PCI-E interface and NVMe host protocol.

    On the enterprise side, there is great deal of deployed infrastructure built around the SAS interface (a super-set of SATA), and the small form factor (SFF) for 2.5in HDD at 15mm height or thickness. The bean counter types would like to wish that SSD (NAND flash for those who do not like the acronym SSD) would use existing infrastructure and not just as an interim measure. They are probably still unhappy that Fiber Channel on the back-end had to be abandoned several years ago, being not competitive and a cost burden relative to SAS.

    Preserving the value of investment in existing infrastructure is important because people are unhappy when equipment purchased at painfully high cost becomes obsolete. Of course, enterprise storage is only extremely expensive because storage vendors invented justifications for selling inexpensive components at very high markup. There is also a failure to consider that hardware has an effective depreciation of 30-40% per year due to the pace of progress, which renders the term investment in hardware foolish, or if I were less polite, then completely stupid. So ultimately this is circular logic based on an untenable premise.

    That said, it would be possible to build a viable enterprise storage system around either the PCI-E or SAS interface, because both support multiplexing lanes, and there are switch chips for both PCI-E and SAS interfaces. The reason PCs are moving from SATA to PCI-E is that the NAND interface bandwidth is advancing at a faster pace than any single lane connection can support and SATA does not support multiplexing. (In websites catering to desktop hardware, some say that PCI-E is superior to SATA. This is rubbish by writers with poor technical knowledge. The only important fact is whether the interface supports multiplexing.)

    The reason existing enterprise infrastructure should be abandoned is not because of any deficiency in SAS, but rather that it is built around four lane (x4) uplink and downlink ports. SAS at 12 Gbits/s would only support 4.4GB/s net bandwidth. This might seem to be high because enterprise storage vendors sell crappy systems with pathetic bandwidth capability. The other reason is that most existing infrastructure are either the 24-25 bay SFF in 2U or 15-bay LFF in 3U 19-inch wide rack mount enclosures designed for hard disks. Both the number of bays and physical volume are completely wrong for current generation SSDs going forward.

    My opinion is that the correct uplink and downlink for solid state (be it NAND flash, Intel 3D XPoint or other) storage enclosures (not individual devices) should be 16 lanes wide or x16. Both PCI-E and SAS have adequate bandwidth and protocols. For PCI-E gen 3 at 8Gbit/s per lane, this would support a net bandwidth of 12.8GB/s. The existing x4 SAS is just too low for an SSD (formerly disk) array enclosure.

    The core of this argument is based around the PC standard of a Flash controller with 8-channels on the NAND side, and PCI-E on the uplink side. Today the NAND interface is 333MB/s, so an 8-channel controller could support 2.6GB/s. There may have been some thought that the upstream side should be PCI-E gen 3 with 2 lanes (x2, capable of 1.6GB/s), as it is common to have excess bandwidth capability on the down stream side. But in PC world, single device benchmark performance is important, so the trend seems to be PCI-E x4 on the controller, with the option to connect only x2 (or even x1?).

    In the time of hard disks, client side PCs used 7200 RPM HDDs or less, for lower cost and higher capacity. Enterprise storage was primarily 10K or 15K RPM for greater IOPS performance, although 7200 RPM was adopted for tier 2 storage. (Storage capacity should have been too cheap to meter even for 10K HDDs, but because vendors sold a ridiculously high prices, this created a demand for 7.2K in enterprise storage systems.)

    In the first phase of SSD adoption, enterprise systems preferred single level cell (SLC) NAND with greater write endurance while client side was mostly 2-bit MLC and later some low cost devices being 3-bit TLC. Today NAND flash technology is sufficiently mature that MLC has sufficient write endurance for many enterprise needs. Fundamentally, the performance oriented PC and enterprise could use the same SSD, just with different over-provisioning and other firmware settings. It would be foolish for enterprise systems not to leverage components developed for client side systems, given the huge volume and low cost structure.

    While the standard desktop SSD element is M.2 form factor with an 8-channel controller and capable of x4 on the upstream side, the enterprise strategy should be to connect x2 on the upstream side. In enterprise, it is the performance of the complete array of storage elements that is important, not the single component. The standard storage array enclosure should probably have 16 bays, each connected x2 to the PCI-E switch, and x16 for each of the uplink port and downlink expansion port. The PCI-E switch would have 64 ports, 16 for uplink, 16 for downlink expansion, and 16 x2 for the M.2 SSDs. The enclosure should work with either 1 or 2 controllers. Existing DAEs have a single (x1) SAS connection to each bay.

    The physical volume for 16 M.2 devices would occupy only one-quarter of 1U rack. Existing enterprise storage infrastructure is x4 uplink/downlink ports, 2U full rack with 24-25 bays connected x1. This wrong for SSDs at on multiple points. Uplink and down link ports should be x16. The volume of the enclosure should be shrunk by a factor of 8. Connections to each bay should be x2, but 16 bays connected at x1 is weakly viable. Given that existing infrastructure is unsuitable for SSDs going forward, there is no excuse to not adopt the client-side components with M.2 form factor and PCI-E in a new properly designed infrastructure.

    Addendum
    for some reason I cannot respond to comments

    Good question. Sorry about the long winded answer, but life is complicated. I do agree with shared storage concept in HDD days, having a common pool of HDD so that each host can access the aggregate IOPS capability when needed. This and managing the complex storage system alone would have justified a good profit margin. But storage felt the need to justify extraordinary margins, hence started to invent reasons, which led to doctrine based on the invented justifications. Any time happens, it is a fuck-up of the first magnitude. And storage vendors do not seem to understand what bandwidth is, or about log write latency.

    Next, blade systems are non-starter in databases because it gives up DIMM slots and PCI-E slots. So we should stick with rack systems with the full boat of DIMM and PCI-E slots. Today a 4TB PCI-E is do-able. Whats missing is some way to match PCI-E SSDs to the available PCI-E lanes. System vendors have a mix of PCI-E slots, including several x16. Only workstations and HPC have x16 cards, servers do not. So we want to connect four PCI-E x4 SSDs to x16 slots. HP workstations have some card for this, but we need a server version. I can see a 4-socket server with 64 - 128 PCI-E lanes dedicated to storage, that is 16-32 x4 PCI-E SSDs, so 64-128TB in one system. All this can be internal, SSDs do take much space, aren't too heavy and don't consume much power.

    Storage vendors still want to sell horribly expensive AFA storage with features that we don't need, and cannot deliver anywhere close to the bandwidth that is possible. So it really is a fight between the DBA for cheap SSD at high bandwidth, and the SAN admin who wants to spend a shit load of money, have control over all storage, make you fill out forms to justify why you need each precious GB of space, all so he/she can deny you request as not sufficient in his/her judgment.

    Edit 2016-Jan-08
    Given that the NVMe controllers seem to be PCI-E x4, perhaps the strategy should be x16 uplink and downlink, with 8 x4 bays. There is still a 2:1 mismatch between downstream and upstream. The point being bandwidth max is reached with 4 devices, but there is space for 4 more. Above, I suggested 16 x2 bays.

  • Computers without ECC memory are crap - no exceptions

    In my previous post, Hardware rant 2015, some readers reacted to my suggestion that vendors start offering the Intel Xeon E3 v5 in laptop and desktop PCs as if this were an outlandish or impractical idea.

    First, doing so requires almost no additional work. Simply substitute 1) the Xeon E3 v5 for the Core i7 gen 6, 2) the server PCH (C236) in place of the desktop PCH (Z170) which are really the same thing as are the two processors, and 3) ECC memory for non-ECC, which has 8 extra bits over the normal 64-bits. The cost of this might one hundred dollars mostly driven by the premium Intel charges, only slightly from the 12% higher cost of memory. (It is the Xeon E5 line that would be seriously impractical in a laptop that an old person could easily carry. A young fit person might claim to not feel the difference between 4 and 15lbs, or 2 and 6kg).

    Second, I should explain why ECC memory is so important, far out weighing the extra cost. This is true for user systems, not just servers with 24/7 requirements. As the title states, a PC without ECC protected memory is total crap, no exceptions unless what you do on the PC is totally worthless, which could be the case for a significant market segment.

    Basically without any check on memory integrity, we may have no idea when and where a soft error has occurred. Perhaps the only hint being the OS or application crashes for no traceable reason or serious data corruption has already occurred. Let it be clear that soft errors do occur unless you are deep under ground.

    Up until the early 1990’s, many if not most PC’s sold as desktops and laptops had parity protected memory. Then in the time frame of Windows 3.x, (almost?) all PC vendors switched to memory with no data integrity protection for their entire lineup of desktop and mobile PCs (with perhaps the exception of dual-processor systems based on Pentium Pro and later, that were subsequently classified as workstations). This was done to reduce cost, eliminating 1/9th of the memory for parity.

    All server systems retained parity, and later switched to ECC memory even though entry level servers use the same processor as desktops (either with the same product name, or different). The implementation of memory protection is done in the memory controller, which was on in the north-bridge in the past, and more recently, integrated into the processor itself (starting with Opteron on the AMD side, and Nehalem in the Intel side).

    I recall that the pathetic (but valid?) excuses given to justify abandoning parity memory protection was that DOS and Windows were so unreliable so as to be responsible for more system crashes than an unprotected memory system. However, since 2003 or so, new PCs were sold with operating system shifted to the Windows NT code base, imaginatively called Windows XP.

    (In theory) Windows NT is supposed to be a hugely more reliable operating system than Windows 3.1/95, depending on the actual third-party kernel mode drivers used. (Lets not sidetrack on this item, and pretend what I just said is really true). By this time, the cost of sufficient DRAM, unprotected or ECC, was no longer as serious a matter, even though base memory configuration had grown from 4MB for Windows 3.1 to 512MB if not 1GB for Windows XP or later. And yet, there was not a peep from PC system vendors on restoring memory protection with ECC now being standard. (I did hear IBM engineers* propose this, but nothing from PC vendors without real engineers. We don’t need to discuss what the gutless wonders in product management thought).

    Presumably soft-errors are now the most common source of faults in systems from Windows NT/XP on. Apple Mac OS (from version?) and Linux are also protected mode operating systems. So this is pretty much the vast majority of systems in use today. It is possible that bugs in drivers from third-party that have not been tested under the vast range of possible system configurations (more so for performance oriented graphics drivers?). Still, the fact that vendors to do not regard correcting the most serious source of errors in PCs today is an indication that they consider the work we do on PCs to be worthless crap, which is the same regard we should have for their products.

    Let me stress again that putting out PCs with ECC memory does not require any technical innovation. ECC capability has been in entry server systems built from identical or comparable components all along. By this time, Intel memory controllers had ECC capability which could be (factory) enabled or disabled depending on the targeted market segment. (Intel does have dumbed-down chipsets for the low-end PCs, but it is unclear if ECC was actually removed from the silicon.)

    Additional notes:
    A. The Wikipedia article ECC memory cites references that mentions actual soft-error rates. There are a wide range of values cited, so I suggest not getting hung up on the exact rate, and treat this as order-of-magnitude(s). There is a separate entry soft-errors for anyone interested in the underlying physics. Of course there are other Wiki entries on the implementation of ECC.

    Briefly, the prevalent source of soft-errors today originating with cosmic rays striking the upper atmosphere, creating a shower of secondary particles, of which neutron can reach down to habitable areas of Earth. Unless the environment is a cave deep underground, there will be soft errors caused by background radiation. The probability of errors also depends on the surface area of memory silicon, so a system with a single DIMM will experience fewer soft errors than system with many DIMMs.

    B. Early memory modules were organized as 8 bit data plus 1 bit for parity in a 30-pin x9 SIMM. Sometime in the early 1990’s, around the 80486 to Pentium time, 72-pin x36 SIMMs (32 bit data, 4 bit parity) was popular. The implementation was 1 parity bit protects 8 bits of data for both the x9 and x36 modules. Parity protected memory had ability to detect, but not correct single bit errors in an 8 bit “line”.

    A few high-end servers in this era had ECC memory which may have been implemented with 2 x36 memory modules forming a 64 bit line with 8 bits for parity, or perhaps a custom memory module? Later on, memory modules progressed to DIMMs, having 64 bits of data with allowance for 8 additional bits for ECC. The base implementation of ECC is to have a 72-bit line with 64-bits for data and 8 bits for ECC. This allows the ability to detect and correct single-bit errors and detect but not correct 2-bit errors (SECDED). More than 2-bits in error could potentially constitute an undetected error (dependent on the actual ECC implementation). There also other ECC strategies such as grouping 4 x72 DIMMs into a line allowing the ability to detect and correct the failure of an entire x4 (or x8?) DRAM chip, when each DIMM is comprised of 18 x4 chips, each chip providing 4 bits of data.

    C. At the hardware level, if an error is detected and corrected, the operating system and applications continue to function. The event can be logged at the system level. A detected but uncorrected error, the hardware should cause a blue screen OS crash.

    An undetected error is just that. It is undetected. The system continues running with incorrect memory content. Depending on the nature of the memory corruption, anything can happen. It could be executable code, in which case the instruction changes. It could be critical a operating system data, causing subsequent memory access to read or write to the wrong location, which could have serious corruption consequences. It could also be end data, or number or char or control, which may or may not be critical.

    Edit
    * It is probably more correct to say that soft-errors is the province of scientists/physicists, not engineers. Sun had perfectly good engineers, but in the late 1990's, they had an Ultra-Sparc II processor with 4M L2 cache in their high-end enterprise system. I believe the L2 data had ECC - SECDED, but the tags were only parity protected - SED. Some of systems started to experience mysterious failures (the one located in high-altitude locations?). This was ultimately traced to soft-errors. It was not a simple thing to change the L2 cache tags from parity to ECC (logic in the processor itself?) so the temporary solution was to mirror the memory used for tags? (if some knows the details, please step forward)

    Edit 2015-11-10
    The Wikipedia topic ECC Memory states "ECC memory is used in most computers where data corruption cannot be tolerated under any circumstances, such as for scientific or financial computing."
    It is more correct to say ECC is used to when it is necessary to correct the more common (typically single bit) errors, and detect certain errors involving more than 1 bit, which cannot be corrected. However it is possible that some multi-bit errors cannot even be detected.

    Edit 2015-12-10
    Donaldvc pointed to this new article on IEEE Spectrum drams-damning-defects-and-how-they-cripple-computers
    Much of my knowledge is very old, from back in the days when memory chips were 1-4 bit wide. Back then, the soft-error might only affect many memory cells but it would only be one bit in a word. Then as memory became more dense, a soft error could affect multiple bits in a word? So processors did ECC on a bank of 4 DIMMs = 256 bits of data, 288 bits of memory, which allowed more sophisticated algorithms. I am not sure what Xeon E3 or E5 has. Xeon E7 is supposed to be very sophisticated. If someone free time, please look into this.

  • Hardware rant 2015

    It has been a while so I suppose it is time for another rant on hardware. There are two systems I would like:
    One is a laptop.
    The second is a server capable of demonstrating extreme IO performance, with the secondary objective of being small enough to bring along to customer sites.

    On the laptop I am looking for
    1) quad-core with HT, i.e. 8 logical processors for better parallel execution plans.
    2) PCIe SSD, I would prefer 3GB/s+, so PCIe gen3 x4, or 2 x M.2 PCIe x2 is also an option.
    3) drive 1, but preferably 2 external 4K monitors (so I can look at complex execution plans)

    On this matter, it is time to bitch at the MS SQL Server team that there should be an option to contract the white space in execution plans. The existing zoom capability is worthless.
    Yes I know SQL Sentry Plan Explorer can do this, but really MS, is it so hard? or have you outsourced the entire SSMS some team that does not know that there is such a thing as complex queries?
    The reason I want to drive 2 external 4K displays is that at the 4K resolution, I need more than a 28 in monitor to use the resolution.

    A couple of days ago, Dell announce the new XPS 15 with Core i7-6700 processors (Sky Lake) which I immediately ordered , but unfortunately it shows a shipping date of Nov 16

    it does have 4K display, and 1 external port which may or may not support 4K. I thought I ordered the docking station, but I do not know if this would support dual external 4K monitors.
    I currently have the Dell 28in 4K monitors, which is great for looking at execution plans, but at the normal text size setting, is difficult read.
    I am thinking that the much more expensive Dell 32in 4K monitor will be better, but maybe not enough. Should I get a 55in 4k TV instead? these all have just the HDMI connector, so I need to make sure there are proper 4K adapters.

    The new XPS 15 data sheet says it has HDD bay (SATA interface) and one M.2 bay (uncertain if PCIe x2 or x4). I would have been nice if 2 M.2 x2 bays were available instead of the HDD bay. I ordered the XPS 15 with the PCIe SSD. I do not know if it is good one (Samsung SM951 cite 2150MB/s) if not, it will throw the Dell SSD out, and get a good one.

    One more thing, ECC memory
    Intel desktop and mobile processors all do not have ECC (or parity) memory capability. ECC memory has been built into Intel processors for some time now, it is just disabled in the Core product lines, enabled on in the server Xeon line.
    So the Sky Lake equivalent is the Xeon E3 v5. Intel released the v5 under the mobile group, with a 45W rating.
    Unfortunately I cannot find a laptop for sale that uses the Xeon E3 v5.

    Perhaps Dell or someone could offer a Xeon E3 mobile system?

    Extreme IO Performance demonstrator
    First, why do I need such a thing?
    when my clients have multi-million dollar SAN storage systems?
    Because SAN people are complete idiots on the matter of IO performance, being locked into irrelevant matters (to enterprise DB) like thin provisioning etc.
    Invariably, the SAN people (vendor sales engineer, the SAN admin etc) confuse that Fiber Channel is specified in Gigabits/sec (Gb/s) while all other IO bandwidth is specified in GigaBytes/sec (GB/s).
    So we have a multi-million dollar storage system (full of add-on software that have no purpose in an enterprise DB) connected to a powerful server (60+ cores and paying for SQL Server 2012 EE per core licensing) over 2 x 8Gbit/s FC links.
    Is this stupid or is this exceptionally stupid?

    Yes I know it is extremely rude of me to call other people stupid, and that being stupid is not crime, but when you are the vendor for multi-million dollar equipment, there is a reasonable expectation that you are not stupid.

    So onto the system.
    For this, I am sure I need more than 4 cores, so it needs to the Xeon E5. Perhaps 8 cores (single socket) is sufficient.
    The new Intel SSD DC P3608 has great specs, but I am not sure when it is actually available?
    I would put 2 of these in the system to demonstrate 10GB/s. Ideally this would all go into box that fits carry on luggage, which is unfortunately not one of the standard PC or server form factors.

    Another option is a 2 x 12 core system to demonstrate 20GB/s on 4 x P3608.

    Addendum
    I would prefer to get a laptop without high performance graphics, the NVidia GTX 960M in this case.
    The current Intel graphics is sufficient for high resolution rendering, but I do not need high frame rate. All the Intel Core i7 6th gen processors have graphics, I wonder if I can remove the GTX (for power savings)?
    Apparently Dell will have a new docking station, the Thunderbolt Dock TB15 next year, that will support 2 x 4K monitors?

    2015-10-20
    I did already rant on PC laptops only being available with 16x9 displays?
    How stupid is this? It is one thing for consumer laptops to have a 16x9 display, on the assumption that the home users just watch movies.
    but on what justification does this apply to business and workstation laptops?

    Concurrent with the Intel Skylake Xeon E3 v5 regular announcement, Supermicro announced motherboards for the E3 v5.
    There is a micro-ATX (X11SAE-M) but with just 1 x16 and 1 x4 PCIe g3 slots.
    where as the ATX (X11SAT) has 3 slot with 16/8/8 as an option. This would let me put 2 P3608? for 10GB/s?

  • GHz and MHz-BIOS updates and Amdahl Revisited

    Last week, a routine performance test ran about twice as long as expected. A check of dm_exec_query_stats showed that CPU bound statements (worker time roughly equal to elapsed time) were approximately 3 times higher than previous tests for matching SQL statements. Almost all of the SQL involved single or few row index seeks, usually Select, some Insert and Update. The server system is a 2-socket Xeon E5-2680 (8 cores, 2.7GHz nominal, Sandy Bridge) in a managed data center. The data center had sent out notice that there would be system restarts the previous weekend, which could mean either OS patches or BIOS/UEFI updates. So naturally the next thing to do is check the Processor Information object for the Processor Frequency and % of Maximum Frequency counters (or equivalent). This showed 135, as in 135MHz, and 5% of maximum. Another system of the same model also rebooted showed 1188 and 44%.

    This issue has occurred previously in this environment and in other HP systems that I am aware of. The BIOS (or UEFI) update puts the system into one of the energy efficient configurations. It could also be an operating system setting, but most that I have seen are BIOS settings? One can imagine a firmware engineer being so committed to green activism that this was made the default on BIOS updates without discussion with other parties. Perhaps there is a facility (in Houston?) with inadequate air conditioning for the number systems, that this setting was put in to prevent the lab from overheating. Then no one remembered to exclude the step in the production code? Not that I have ever done such a thing (and no further questions on this should be asked).

    Another question might be why the data center monitoring team did not check for this, as it has happened before. The whole argument for going to managed data center instead of a simple hosted data center was that the managed data center could provide the broad range of expertise that is not economical for a mid-size IT department. Obviously this managed data center did not monitor for the performance/power configuration.

    This matter is of serious concern to production DBAs and IT staff in handling operations. As the Processor Information performance object with extended information was only introduced in Windows Server 2008 R2, many software monitoring tools may not alert on changes of Processor Frequency, especially after reboot. Imagine the IT staff or DBA encountering this for the first time on the production servers, with users complaining, your boss watching over your shoulder, and his/her boss hovering over your boss, offering their helpful insights in the non-judgemental manner as bosses do.

    Performance Insight

    However, I am more interested in a different aspect of this incident. When there are two sets of data, one for the processors cores at 2.7GHz and another at presumably 135MHz, we can extrapolate parameters of interest. Does it seem stunning that the drop from 2.7GHz to 135MHz, a factor of 20, only decreases CPU efficiency (increase CPU-sec, or worker time) by 3X? Perhaps, but this actually should have been expected.

    The salient aspect of modern computer system architecture is the difference between CPU clock cycle and memory access time. A young person might not know, but old timers would know. Up to about 20 years ago, the primary memory performance specification was access time, with 80, 70 and 60 ns being common in fast page mode and extended data out. Then with the switch to synchronous dram (SDRAM), the key specification changed to data rate. In the Xeon E5 (v1) generation, DDR at 1333MHz was common. This means a memory channel can deliver one line every 0.75ns, or 1.333 billion times per sec, with a line being 64-bits (excluding ECC bits). The Xeon E5 26xx series has four DDR3 channels. The Intel processor internally is shown as having 2 memory controllers, each controller driving 2 DDR channels, so channel can have different meanings depending on the context).

    What is less commonly cited is the round trip latency, from a processor issuing a request to memory, the internal memory access within the DRAM chip and finally the transmit time back to the processor. (The L1, L2 and L3 cache sequence is also involve in memory access timing.) For local memory (attached directly to the processor) this is around 50ns. For memory on an adjacent processor, the round trip time might be 95ns or so.

    On a 2.7GHz processor, the CPU cycle time is 0.37 ns, so 50ns for local memory round trip access is 135 CPU cycles. This particular system has 2 sockets, so one might expect that half of memory accesses are local at 50ns round-trip latency, and half at 95ns latency.

    This is a well understood issue. Two methods of addressing the disparity between CPU cycle time and memory access are 1) large cache on the processor, and 2) pre-fetching memory. Current Intel processors have dedicated 32KB I+D L1 and 256K L2, both per core, and an additional shared L3 cache sized at 2.5MB per core. From Pentium 4 one, the processor pre-fetches 64-bytes (the cache line size) with an option to prefetch the adjacent cache line. Prefetching is exposed in the instruction set architecture (can someone provide a reference please) and there should also be a BIOS/UEFI for hardware prefetch.

    Simple Model

    Now lets visualize the (simplified) code sequence in a relational database engine with traditional page-row data structures. There is a memory access for the index root level page. Read to the page to find the pointer for the second level page. Memory access, and repeat. It is a sequence of serialized memory accesses with poor locality (so cache can only help so much) and the next location is not known until the current memory request is completed, so pre-fetching is not possible.

    Modern processor performance characteristics are very complicated, but we will attempt to build a very simple model focusing on the impact of round-trip memory access latency. Start with an imaginary processor with a cycle time equal to the full round-trip memory access time. In this scenario, one instruction completes every cycle, be it an arithmetic or logic or memory access instruction.

    Such a system may have never existed so now consider a system where the round trip memory access latency is some multiple of the CPU cycle time. The average time to complete an instruction where time is in units of the memory access latency (50ns or 20MHz for local node), “a” is the fraction of instructions that involve (non-local, non-prefetch-able) memory access and “n” is the processor frequency.

      (1-a)/n + a

    The term (1-a) is the fraction of instructions that are either not memory access, or memory access to cache (from previous access or pre-fetched). “1/n” is the processor cycle time (in units where memory access time is 1).

    Performance (instructions per unit time), the inverse of average time per instruction is:

      P = 1 / ( (1-a)/n + a )

         = n / (1 +(n-1)*a )

    We can see the the above equation has characteristics that as processor frequency increases, the upper bound on performance is:

      n -> infinity, P = 1/a

    Also, if the fraction on instructions that require memory access, “a,” is zero, then P = n.

    Does the above look familiar? It is just Amdahl’s Law, which formulated in the old days to demonstrate the limits of vectorization in supercomputers. I have just recast it to express the limits of increasing processor frequency relative to round-trip memory access time.

    If someone would like to check my math, please do so. It has been a long time. Trying tricking your teenage son/daughter into doing this as a practical math exercise?

    OK, anybody still reading is obviously not deterred by math, or knows the trick of skipping such things. What am I really after? In the above equation, what is known is processor frequency relative to memory access latency. While we know the performance or worker time of certain queries, we do not know it terms of instructions per CPU-cycle. And the second item we do not know is the fraction of instructions that introduce a round-trip memory access latency that cannot be hidden with cache or pre-fetching.

    But, we have data points at 2 frequencies, 2.7GHz and reportedly 135MHz. Express the relative performance between the two points as a ratio.

      P2/P1 = R

    Then from the two equations

      P1 = 1 / ( (1-a)/n1 + a )

      P2 = 1 / ( (1-a)/n2 + a )

    we can solve for a in terms of the know values n1, n2 and R.

      a = (n2 – n1*R) / ( n1*n2*(R-1) + n2-n1*R )

    Assuming memory access latency of 50ns, the base frequency is 20MHz corresponds to memory access in 1 cycle. Plugging in the values n1 = 135MHz / 20MHz = 6.75, n2 = 2700/20 = 135 and R = 3. We get a = 0.059, or 5.9% of instructions incurring a non-cached, non-prefetch round-trip memory access latency would result in a 3:1 performance ratio between 135MHz and 2700MHz. (Perhaps it would be more correct in estimating round-trip memory access latency as the average between the local and 1-hop remote node at 75ns?)

    So while it might seem astonishing that the difference between 135MHz and 2700MHz translates to only 3X performance, the database transaction processing workload is an extreme (but important) case. There are many workloads which exhibit better locality or have memory access patterns that are amenable to prefetch and have performance scaling better with frequency.

    Hyper-Threading

    Earlier, two methods of hiding round-trip memory access latency were mentioned. There is another, Hyper-threading. The processor core to appears as two (or more) logical processors to the operating system. Presumably, there is an extra set of program counters, and resources to determine which physical registers (different from the registers specified in the instruction set architecture) are assigned to each logical processor.

    In the earlier example, say that the round-trip memory access time is 135 CPU-cycles and the fraction of instructions that incurs the full round-trip latency is 6%. Then for 100 instructions, 94 are executed in 1-cycle each (excluding consideration for superscalar) as either not involving memory access or data is already in cache, and the 6 the incurs the round-trip memory latency of 135 cycles. Then the total time in terms of CPU-cycles is 94*1 + 6*135 = 904. In other words, only 100 cycles out of 904 are used, the rest are no-ops.

    The Intel Xeon processors from Nehalem on implement Hyper-Threading with 2 logical processors on each physical core. (This can be disabled in BIOS/UEFI. Some models have HT disabled. The earlier Intel Pentium 4 based processors implemented a more complex form of HT.)

    In considering the nature of the database transaction processing workload, being a memory access to determine the next memory access in nature, it is perhaps time for Intel to increase the degree of HT, especially considering that the server oriented Xeon E5 and E7 models are already 1 full year or more behind the smaller desktop/mobile processor variants. I seem to recall IBM POWER as having 4 logical processors per physical core, one of the SPARC processor lines as having 8. It would also be necessary to have a good strategy for using HT based on workload. The option to enable or disable HT in the BIOS/UEFI is not I what mean. HT should be visible to the OS. But the application itself should detect the presence and degree of HT, and make its own decision on whether HT should be used and how it should be used.

    Xeon Phi, Many Integrated Core

    Another item worth mentioning here is the Intel many integrated core (MIC) architecture, codename Knights something, now Xeon Phi. The processor puts many smaller processor cores, 61 in the 22nm Knights Corner, versus 12-18 in the 22nm mainline Xeon processors. The theory behind many smaller cores stems from one of the two main elements of Moore's Law. Doubling the number logic transistors/complexity in a single core should translate to about 40% performance gain.

    (This was the case up to several generations ago. Since then, Intel no longer tries to double the logic from one process to the next. There might be 10-20% performance gain in general instructions. Some effort is given to expanding the functionality of the special/vector instructions. And most effort has been in increasing the number of cores.)

    One manifestation of this (more logic transistors) could increased frequency (which Intel stopped pursing years ago). Another might be more execution ports (8 in Haswell) or other areas to improves instructions per cycle (IPC). Following the rule of 2X transistor per 1.4X (square root of 2) backwards, the expectation is that a processor if 1/4th the size would have 1/2 the performance. But potentially there could be 4X as many cores, depending on interconnect and power limitations. So in workloads that are amenable to vectorization, or otherwise can be parallelized, the more smaller cores could be a better strategy.

    The Xeon Phi is targeted to HPC workloads, as reflected in the 512-bit SIMD instructions. If we were thinking about a transaction processing database engine on the MIC architecture, we would probably consider a very basic ALU without SIMD, (not sure on FP). I am thinking that an asymmetric processor architecture might be the objective. Perhaps two powerful cores from the current main line, and many simpler cores (without SIMD) perhaps even simpler than Atom? (The Intel Xeon Phi line implements Hyper-Threading with 4 logical processors per physical core.)

    Column-Store

    As said earlier, the nature of database page storage along rows make serialized memory access (also called pointer chasing code?) its hallmark. This is why there is interest in column storage architecture. Now all of the sudden, for certain database workloads, the next memory access is 4 bytes over, already in cache. The work a little further down touches memory in the next 64 byte line or two away. Both the software and hardware knows this, and either is capable of issuing a pre-fetch. It does not matter that columnstore must touch more data. Processor can stream huge volumes of data, much more effectively than pointer chasing only the necessary rows.

    Summary

    I should probably say something here.

    Notes

    As I said earlier, modern microprocessors are very complex. Pipelined execution was introduced (in Intel processors) with the 80486 (1989) and superscalar execution with Pentium (1993). Pipelined means that while the processor can complete an instruction in each cycle, the actual start to finish time of a specific instruction occurs over several cycles. Intel does not talk about pipeline stages any more, but there are occasional references to Core2 and later processors having a 14+ stage pipeline. (Wikipedia say Core2 is 12-14 stage pipeline. Nehalem and later 20-24?, Sandy Bridge as 14-19.)

    Superscalar means that there are more than one execution unit, with the goal of completing more than one instruction per cycle. Haswell has 8 execution ports. Several processors generation prior were 6-port on superscalar. We could apply the principle of Amdahl’s on scaling performance to any and all of pipe-lining, superscalar, and round-trip memory latency, and probably other things too.

    Rethinking Computer System Architecture

    I have said this else where. It is long past due to do a clean sheet system architecture with matching change to OS architecture. Current system architecture stems from the 1970's of processor with physical memory (8MB was big) and a page file on disk. Why do we still have a page file on disk? In the old days, there was not enough physical memory such that it was tolerable to have a page file on disk to support a larger virtual address space.

    Today, more the 1TB of physical is possible and affordable (compare to the cost of SQL Server per core licensing). But the key factor is in how memory is used. Back then, it was mostly for executable code and internal data structures. The assumption was that very few database data pages would actually be in memory at any given point in time. Today, a very large percentage of memory is used for caching data pages. Of the memory used for executable code and internal data structures, most is junk.

    The CPU-cycle time to memory access time discrepancy dictates that the more urgent strategy is to get memory closer to the processor even if it means drastically reducing the size of true memory, to perhaps a few GB per socket. Given that DRAM is so cheap, we would still have system with multi-TB DRAM capacity, except that this would now be the page file. Of course the operating system (and applications) would have to be designed around this new architecture. Given how well the Intel Itanium software coordination went, I guess this might be too much to expect.

  • Transaction IO Performance on Violin

    Back in Feb, I went on a diatribe-rant against the standard SAN Vendor configuration practice. The Problem with Standard SAN Configuration IO Performance, article and accompanying post, showed IO performance metrics for a batch driven transaction processing workload on a SAN managed by a data center/cloud company. The only option offered by the service provider was to request volumes for storage. There was no consideration for special IO characteristics of transaction processing or other database workloads. No discussion. This practice is doctrine pontificated by SAN vendors, calling SAN admins on a mission to implement the "storage as a service" concept while remaining completely blind to the requirements of mission critical databases.

    Ok, I am venting again. Now I have performance metrics for the same workload, except that storage is on a Violin system. The system is different in having 24 physical cores, no HT, and 256GB memory versus previous system with 16 physical cores, HT (32 logical) and 384GB.

    Below are the IO characteristics. The horizontal axis time scale is 5 min per major division for 1 hour across the entire chart. Each tick is 1 minute. Data points are every 15 sec. Note that the Feb (HDD) charts were 1 min per major division, 15 min total, with data point every 5 sec.

    CPU
    CPU

    tps
    Transactions/sec (red)

    IOPS
    IOPS - read (green), write (blue), log write (red)

    msIO
    ms/Rd or Wr

    The obvious difference between the IO characteristics on Violin and the previous HDD-based storage is that checkpoints now have almost no visible impact on performance. Both CPU and transactions/sec are very steady, with slightly noticeable blips, versus the severe drops before. It is evident that checkpoint writes now have almost no impact on data reads or log write IOPS. The same is true of IO latency, in milli-seconds per read or write.

    If the storage had been on HDD storage but with logs on a separate physical disks, we would expect that the checkpoint would drive up data read latency, and hence depress data read performance. But it would have no impact on log write latency, and hence no impact on log write performance. The lower data reads should have only moderately depress performance, not severely.

    The difference in system processor, 24 physical cores no-HT versus 16 cores plus HT is not a factor in the IO characteristics. The difference in physical memory, 256 GB on the system Violin storage and 384 GB in the system with HDD storage is evident in the data read IOPS, starting at 7-8K IOPS then drifting down to 2-3K IOPS on the system with less memory, compare with mostly 1K IOPS on the system with more memory. Both storage systems can easily handle 20K IOPS.

    The main argument here is not that SSD/Flash storage is a requirement for transaction processing databases, even though there are clear benefits. (NAND flash based SSD have both maturity and cost-structure to be very attractive for any new storage system purchases.) The point is that there is a severe problem with the SAN vendor doctrine of one common pool for all volumes.

    This very severe problem can mostly and easily be mitigated simply by having separate physical disks for the log volume. So the point could and should be demonstrated by showing the IO performance on an HDD SAN with separate physical disks for logs. But this violates the SAN doctrine of ignoring user requirements, and would not be considered or allowed by the service provider under any circumstance. So the only real solution is the keep performance critical databases off a storage system administered by a team on a different mission than supporting the database.

    Below are excerpts and the graphs from the Feb article.

    Standard SAN Configuration IO Performance 2015-02

    The chart below is CPU. The horizontal axis is time. One major division marked by the vertical line is 1 minute, and the small tick is 12 sec. The data points are 5 sec. There are 12 steps between each major division. The vertical axis is overall (system) CPU utilization in percent. Each of the stacked green lines represents an individual processor. There are 16 physical cores and 32 logical. A single logical core at 100% utilization would show a separation of 3.125% to the line below.

    CPU
    CPU

    On the second chart, the red line is the performance monitor object: SQL Server:Databases, counter: Transactions/sec. Note that the vertical axis is log-scale, base 10. One major division is a factor of 10. Each minor tick is an integer. The first small tick up from a major tick is 2, the next is 3 and so on to 9 for the last small tick.

    transactions/sec
    time scale
    Transactions/sec

    The third chart is IOPS. Green is data reads, blue is data writes, and red is log writes. The vertical axis is log scale.

    IOPS
    IOPS

    The fourth chart is IO latency, milli-sec per IO. The same color codes applies as for IOPS. Again the vertical axis is log scale.

    IO latency ms
    ms/Rd or Wr

  • SAN Configuration Performance Problems

    The charts provided here illustrates my complaints against SAN vendor doctrine, obstinately adhering to the concept of one large pool of disks from which all volumes are created for any purpose (data, log, and junk non-DB stuff). There is no consideration for the radically different characteristics of hard disks in random versus sequential IO (low vs. high queue depth IO behavior should also be an element of IO strategy). The architecture of all traditional relational database engines are built on the premise that high volume log writes are possible at very low latency (using dedicated disks) in order to provide durability of transactions. And yet SAN vendors blithely disregard this (because it is at odds with the doctrine derived from principles invented) to justify their mission to sell inexpensive commodity hardware components at extraordinary prices.

    Transaction Performance Data

    The chart below is CPU. The horizontal axis is time. One major division marked by the vertical line is 1 minute, and the small tick is 12 sec. The data points are 5 sec. There are 12 steps between each major division. The vertical axis is overall (system) CPU utilization in percent. Each of the stacked green lines represents an individual processor. There are 16 physical cores and 32 logical. A single logical core at 100% utilization would show a separation of 3.125% to the line below.

    CPU

    On the second chart, the red line is the performance monitor object: SQL Server:Databases, counter: Transactions/sec. Note that the vertical axis is log-scale, base 10. One major division is a factor of 10. Each minor tick is an integer. The first small tick up from a major tick is 2, the next is 3 and so on to 9 for the last small tick.

    transactions/sec
    time scale

    The third chart is IOPS. Green is data reads, blue is data writes, and red is log writes. The vertical axis is log scale.

    IOPS

    The fourth chart is IO latency, milli-sec per IO. The same color codes applies as for IOPS. Again the vertical axis is log scale.

    IO latency ms

    The SQL

    The SQL activity is batch driven transaction processing. There are 14 or so threads concurrently looping through a set of items to be processed. Each item involves about 20 rows of insert or update activity, hence 1000 log writes per sec corresponds to approximately 20,000 transaction/sec on the SQL counter.

    Most of the active data is in memory. There are probably 30-40 SELECT rows for each transaction or twice as many reads as writes. The data IO reads necessary to support the 20,000 inserts and updates/sec is about 2,000/sec, which the storage system is capable of supporting at about 4ms latency. This is because the data resides a small part of each disk. The actual latency for random IO is less than the expected value of 7.5 ms for data randomly accessed over an entire (10K) disk at queue depth 1.

    For approximately 20 seconds out of each minute, the transaction rate drops from the peak value of 20,000 all the way down to about 8,000 per sec (noting the log scale). These are the check points when the data write IO surges to 20-50K IOPS, (which demonstrates that the storage system is actually pretty decent) though write latency is driven up to 50-90ms.

    The checkpoint surge also pushes log write latency up from 1ms to 20-30ms. This dataset occurred during the day, when presumably there was activity for other hosts on different volumes but on the same SAN. At night, log write latency away from checkpoints could be under 0.3ms even at high volume.

    The Storage System

    I had not previously discussed the storage configuration in detail. The storage system consists of 240 x 10K HDDs only, with the standard system level caching. The SQL Server host is connected to the SAN over 4 FC ports (8Gb from host to switch, 4Gb from switch to SAN, and presumably 4Gb on the SAN backend?). The data is distributed over 8 volumes. The log is on a separate volume as seen by the host OS.

    The Problem

    The problem is that on the SAN, all disks are aggregated into a single pool, from which volumes are created. This was done per standard SAN vendor doctrine. Their magically great and powerful "value-add" intelligence would solve all your performance problems. We cannot ask for dedicated physical disks for the log because the SAN was already configured, with the SAN admin getting assistance from the SAN vendor's field engineer who followed the SAN vendor's doctrine.

    Input from the DBA team was not solicited and would have ignored in any case. Besides, there are no unallocated disks. And no, the SAN team will buy more disks because there are no empty bays in the disk enclosures. And there is no room for more enclosures in the storage rack. So the DBA request is denied.

    Even if we put up the money to get an extra cabinet for one more disk enclosure, the SAN admin will still refuse to configure dedicated physical disks for the log volume because the SAN vendor said that their great and powerful SAN will solve all performance problems. Any problems must be with their application and not the SAN. As can be seen from the charts above, this is a load of crap.

    The SAN Vendor Solution

    As I said above, this particular SAN is comprised of 240 or so 10K HDDs.

    Naturally the SAN vendor's proposed solution is that we should buy more of their value-add products in the form of auto-tiering SSD-HDD, and perhaps additional SSDs as flash-cache. The particular SAN with base features probably has an amortized cost of $4,000 per HDD. So SAN with 240 disks would cost just under $1M (while still failing to provide desired database performance characteristics). A mid-range SAN might have amortized cost per disk of $1,500-2K. Enterprise SAN could be $4-6K per disk.

    The additional value-add features would substantially increase the already high cost, while providing only minor improvement, because the checkpoint IO surge will still drive up log write latency.

    The sad thing is that the SAN group might buy into this totally stupid idea, and refuse to acknowledge that the correct solution is to simply have dedicated physical disks for the logs. If there were dedicated physical disks, the checkpoint data IO surge goes to completely different physical disks than the log disks.

    In the specific example, it is not necessary to have separate FC ports for the logs. The 50K IOPS at 8K per IO generates 400MB/sec, which is only 25-33% of the realizable IO bandwidth of 4 x 4Gbit/s FC ports. The checkpoint IO surge would increase latency on data reads, but the primary reason for the performance drop is the increase (degradation) in log write latency.

    Another angle is changing the checkpoint parameters in SQL Server, but the real problem is because we are prevented from leveraging the pure sequential IO characteristics of HDDs by allocating data and log volumes from a common pool.

    One more item. In the old days before we had immense memory, typical transactional database data read/write mix was 50/50. This is because a read forces a dirty page to be written to disk. In this situation, a data write IO surge would also depress the data reads necessary to support transactions. So the standard practice those days was to set the checkpoint interval to infinity to prevent data write IO surges. In our case, very little data reads are necessary to support transactions, so the checkpoint surge might depress data reads should have lesser impact on transactions. It is the increase in log write latency that is depressing transaction volume.

    Solutions that work

    A solution that would work is simply to have separate dedicated physical disks for the log volume. It is that simple! And yet this is not possible because the SAN people would refuse to do this, as it is not in their agenda.

    It is unfortunate that the only practical solution is to get the critical database off the corporate SAN. I have done this by going to all flash in the form of PCI-E SSDs. That is, SSDs installed internal to SQL Server system. This is not because the exceptional performance characteristics of SSDs were required.

    It was because I needed to get away from the SAN admin and his dogmatic adherence of SAN vendor doctrine. The IO performance requirements could have been meet with a direct-attach HDD array (or on a SAN). But anything with HDD enclosures would have been under the authority of the SAN admin, who would have nixed any storage components that was not a SAN, and then configured it according the SAN vendor principles.

    I have used the excuse that PCI-E SSD "accelerators" are needed for tempdb, which are not really "storage", hence there is no IT department mandate that it be on the SAN, under the absolute control of the SAN admin. In fact there were no special requirements for tempdb different from that of data. Then for unrelated reasons, there was enough capacity to put the entire DB on local SSD. Perhaps a file group with non-critical objects could reside on the SAN to carry the pretense of the local SSD not really being storage.

    Addendum

    Note that I have not naively suggested engaging in constructive dialog with the SAN team. They are on a holy mission that has no alignment with their company/organization's mission. Anything that contradicts SAN scripture is heresy.

    Oracle database machine has been described as hardware optimized for Oracle database. In fact the true mission is to take the SAN admin out of the loop. I think HP offered appliance oriented systems for SQL Server in 2012?, but they only system option was the DL980? which is severely narrow market segment. There needs to be DL380 and 580 options as well.

    6-Mar

    A SAN is simply a computer system that resides between storage elements and hosts (storage clients) providing volumes (similar to any other client-server application). One practical feature is that the SAN can provide protection against single component failure. I have no objection to this concept, nor to the fact that the SAN vendor wants to sell hardware at a list price of 10X markup to cost.

    Strangely, it seems that people want to buy expensive stuff (just look around outside of IT). Consider that an organization might spend anywhere from tens of millions to a few billion dollars to develop the main database application. It does not seem right (to a high level executive) to put such an application on a $100K server plus $200K storage when there are $1M servers and $6M storage systems available. Never mind whatever might be the consequential differences between them. Or the fact that there are supreme technical challenges in scaling on (hard) NUMA, and oh yeah, the chief architects have never heard of NUMA.

    The point here is that people will buy very expensive SAN systems without technical understanding or justification. There is a perfectly sound business justification. What the client needs is a system supported by deep experts. Not some rookie field engineer who incorrectly chooses to replace the memory board with no uncorrected errors before then board with uncorrected errors.

    My observation is that the SAN Vendors feels a need to have technical justification for selling a storage system with extraordinarily high marks. So the vendors creates reasons. (Technical justification created for marketing requirements tend to have serious logical flaws, but a creative marketing professional is not easily deterred).

    It follows on pretending that these technical justifications are valid, then the "best" practices for employing a SAN should be blah, blah, as derived from the underlying reason. Do I need to explain the consequences of implementing practices built on a lie?

    I will add that there is an absolutely critical value-add that the storage system vendor must provide that alone justifies very expensive pricing. This is integration testing, verifying that the complex system with very many components work well together. The hard disk vendors are generally good at validating that their products works to specification in desktop and workstation configuration with 1 or even a few drives. Ensuring a storage system with several controllers, thousands of HDD, dual-controllers etc., is vastly more complicated. This is a valid reason. Building practices on a valid reason has benefits.

    8-Mar

    In a large environment, there might be hundreds or thousands of servers, database instances, SQL Server or other. Managing very many systems and databases economically requires a high degree of automation, usually implemented with standard configurations.

    However, there are some databases that are critical either in being required for day to day operations and possibly providing a competitive advantage over other options. One indicator of this is that there are DBAs and developers dedicated to a specific application. In this case, the argument that customization of storage configuration is not feasible because of the other responsibilities on the SAN team is total BS.

  • ExecStats update - automating execution plan analysis

    It has been over 7 years now that I have made my ExecStats
    (current version Exec Stats 2015-02-18)
    tool publicly available (late 2007), with prototype versions going back to 2006. The two distinguishing elements of ExecStats is 1) the emphasis on cross-referencing a) query execution statistics to b) index usage via the c) execution plans, and 2) saving information locally so that it can be sent to a remote expert.

    The more recent versions now have the capability to simultaneous group stored procedure statements together while consolidating SQL by query hash. On the performance (counter) monitoring, charts are now displayed on logarithmic scale, allow insight over several decades of dynamic range.

    Too many of the commercial SQL Server performance tool emphasis generating pretty charts and reports on query execution statistics, thinking that by giving the top SQL to the originating developer is all that is expected of the DBA. In principle the DBA should have greater expertise on SQL Server performance than a developer, who is expert on the development platform (Visual Studio or other) and perhaps the business logic.

    In any case, the query execution statistics by itself is not very helpful, without the execution plan, and possibly the difference between estimated and actual rows. We should first appreciate that SQL Server is of some complexity, with several factors that could have adverse impact on performance.

    Examples are 1) does the table architecture support the business logic, 2) is the SQL written in manner that the query optimizer can interpret in the desired manner, 3) having a few good indexes, 4) no more indexes than necessary, 5) a good statistics strategy, not just frequency of rebuilds but also whether full scan samples are necessary, and possibly explicit statistics update in between certain SQL statements typically for ETL, 6) compile parameter issues, 7) row estimation issues at the source, 8) serious row estimate errors after intermediate operations (not at the source) and of course 9) system level issues.

    Some very advanced tools have the capability of generating alternative SQL in an attempt to solve a performance problem. The question is have other factors been considered before jumping into re-writing the SQL?

    I have had good success with ExecStats in onsite engagements in greatly simplifying data collection and to a degree the analysis as well. On many occasions I have been able solve simple or medium complexity issues remotely with just the information collected by ExecStats. It should be able to work against Azure as well, but this is only tested intermittently (and on request). People are invite to use and send feedback.

    ExecStats2015
    ExecStats documentation

  • extensions for sp_helpindex

    The system stored procedures in SQL Server from the very beginning(?) provide useful information. However they have not been updated in a substantial manner for the new features in later versions, nor have they been extended with additional details that are now available in the DMVs for more sophisticated DBA's. Presumably, this is for backward compatibility. As an alternative, there is a provision for creating custom procedures that behave as system procedure with sp_ms_marksystemobject.

    In the case of sp_helpindex new features of are included columns, filtered indexes, compression and partitioning. Other information of interest might be size, index usage and statistics. Part of the reason for not changing could be for backward compatibility, which is fine, but let's then make new system procedures with extended information.

    In the text of sp_helpindex for SQL Server 2012, the only difference from earlier versions is that the description field has a provision for columnstore. The SQL Server 2014 version adds hash index and memory optimized in the description field.

     

    Below is my new procedure for extended index information. I have retained the same error checking code from the original. The cursor loop to assemble the index key columns has been replaced with a code sequence using the STUFF function and FOR XML PATH. A similar structure reports on the included columns. This procedure does not replicate the index description field of the original, but rather has a limited description and a separate field for the type code.

     

    USE master

    GO

    CREATE procedure [dbo].[sp_helpindex3]

     @objname nvarchar(776)

    AS

    DECLARE @objid int

     , @dbname sysname

     -- Check to see that the object names are local to the current database.

    select @dbname = parsename(@objname,3)

    if @dbname is null

      select @dbname = db_name()

    else if @dbname <> db_name()

    begin

      raiserror(15250,-1,-1) 

      return (1)

    end

    -- Check to see the the table exists and initialize @objid.

     select @objid = object_id(@objname) 

     if @objid is NULL 

     begin 

      raiserror(15009,-1,-1,@objname,@dbname) 

      return (1) 

     end 

     

    ;WITH b AS (

      SELECT d.object_id, d.index_id, part = COUNT(*)

      , reserved = 8*SUM(d.reserved_page_count)

      , used = 8*SUM(d.used_page_count )

      , in_row_data = 8*SUM(d.in_row_data_page_count)

      , lob_used = 8*SUM(d.lob_used_page_count)

      , overflow = 8*SUM( d.row_overflow_used_page_count)

      , row_count = SUM(row_count)

      , notcompressed = SUM(CASE data_compression WHEN 0 THEN 1 ELSE 0 END)

      , compressed = SUM(CASE data_compression WHEN 0 THEN 0 ELSE 1 END) -- change to 0 for SQL Server 2005

      FROM sys.dm_db_partition_stats d WITH(NOLOCK)

      INNER JOIN sys.partitions r WITH(NOLOCK) ON r.partition_id = d.partition_id

      GROUP BY d.object_id, d.index_id

    ), j AS (

      SELECT j.object_id, j.index_id, j.key_ordinal, c.column_id, c.name,is_descending_key

      FROM sys.index_columns j

      INNER JOIN sys.columns c ON c.object_id = j.object_id AND c.column_id = j.column_id

    )

    SELECT ISNULL(i.name, '') [index]

    , ISNULL(STUFF(( SELECT ', ' + name + CASE is_descending_key WHEN 1 THEN '-' ELSE '' END

       FROM j WHERE j.object_id = i.object_id AND j.index_id = i.index_id AND j.key_ordinal >0

       ORDER BY j.key_ordinal FOR XML PATH(''), TYPE, ROOT).value('root[1]','nvarchar(max)'),1,1,'') ,'') as Keys

    , ISNULL(STUFF(( SELECT ', ' + name

       FROM j WHERE j.object_id = i.object_id AND j.index_id = i.index_id AND j.key_ordinal = 0 

       ORDER BY j.column_id FOR XML PATH(''), TYPE, ROOT).value('root[1]','nvarchar(max)'),1,1,'') ,'') as Incl

    , i.index_id

    , CASE WHEN i.is_primary_key = 1 THEN 'PK'

       WHEN i.is_unique_constraint = 1 THEN 'UC'

       WHEN i.is_unique = 1 THEN 'U'

       WHEN i.type = 0 THEN 'heap'

       WHEN i.type = 3 THEN 'X'

       WHEN i.type = 4 THEN 'S'

       ELSE CONVERT(char, i.type) END typ

    , i.data_space_id dsi

    , b.in_row_data in_row , b.overflow ovf , b.lob_used lob

    , b.reserved - b.in_row_data - b.overflow -b.lob_used unu

    , 'ABR' = CASE row_count WHEN 0 THEN 0 ELSE 1024*used/row_count END

    , y.user_seeks, y.user_scans u_scan, y.user_lookups u_look, y.user_updates u_upd

    , b.notcompressed ncm , b.compressed cmp , b.row_count

    , s.rows, s.rows_sampled, s.unfiltered_rows, s.modification_counter mod_ctr, s.steps

    , CONVERT(varchar, s.last_updated,120) updated

    , i.is_disabled dis, i.is_hypothetical hyp, ISNULL(i.filter_definition, '') filt

    FROM sys.objects o

    JOIN sys.indexes i ON i.object_id = o.object_id

    LEFT JOIN b ON b.object_id = i.object_id AND b.index_id = i.index_id

    LEFT JOIN sys.dm_db_index_usage_stats y ON y.object_id = i.object_id AND y.index_id = i.index_id

    AND y.database_id = DB_ID()

    OUTER APPLY sys.dm_db_stats_properties(i.object_id , i.index_id) s

    WHERE o.type IN('U','V')

    AND i.object_id = @objid

    GO

     

    -- Then mark the procedure as a system procedure.

    EXEC sp_ms_marksystemobject 'sp_helpindex3'

    GO

    SELECT NAME,IS_MS_SHIPPED FROM SYS.OBJECTS WHERE NAME = 'sp_helpindex3'

    GO

     

    Information from my extended version of index help are space, index usage, and statistics. The DMV/F function dm_db_stats_properties was added in SQL Server (SQL Server 2008 R2 Service Pack 2, SQL Server 2012 Service Pack 1. The function STATS_DATE was added in SQL Server 2008 can be used if dm_db_stats_properties is not supported.

    We could also join to dm_db_index_operational_stats, dm_db_index_physical_stats or dm_db_xtp_index_stats for additional information, but I do not think the extra information is necessary for routine use.

    The above query might be more useful as a view capable of reporting on indexes for all tables and indexed views, but is there a system view option?

    It is too bad more statistics information in DBCC SHOW_STATISTICS is not available in query form.

    ps
    I am not sure why color coding is not display, see the same page on my web site sp_helpindex3 for
    pps
    We could also do an option to show indexes for all tables, leaving off the object_id =.

  • Join Row Estimation in Query Optimization

    This topic is titled to specifically consider only row estimation after joins, precluding discussion of row estimates at the source table, which has already been addressed in papers covering the new Cardinality Estimator in SQL Server 2014 and other statistics papers for previous versions of SQL Server.

    There are certain situations in which the query compile time can be excessively long even for queries of moderate complexity. This could be true when the plan cost is very high, so that the query optimizer expends more effort to find a better plan before reaching the time out, that is, the time out setting appear to be a function of the plan cost. Even then, the query optimizer could still make row estimation errors in the operations after the source table (for which data distribution statistics are kept on columns and indexes) of sufficient consequence that renders the remainder of the plan un-executable in a practical time frame.

    The new cardinality estimator in SQL Server 2014 is helpful in resolving known issues, but has little to improve row estimates after the initial access at the data source beyond fixed rules which may be more generally true than the rule used before. That said, the query optimizer only attempts to estimate rows (and other cost factors) using a combination of the information it has, and rules for situations where there are no grounds for making a knowledge based estimate.

    So why be bound by this rule of estimating only? The other company (only sometimes in the news concerning databases) has (recently introduced?) adaptive query optimization that can make run-time adjustments to the execution plan. Sure that’s nice, but I am thinking something more sophisticated could done in forming the plan but less complex than the ability to change course in the midst of execution.

    Obviously, if a query has an obviously correct execution plan that can be determined by existing techniques, and the plan cost is fairly low, then no change is necessary. Further special technique should only be considered for expensive queries, particularly in which the row estimates at the intermediate steps are difficult to assess.

    Consider query below
    SELECT bunch of columns
    FROM Nation n
    JOIN Customers c ON c.NationId = n.NationId
    JOIN Orders o ON o.CustumerId = c.CustomerId
    WHERE n.Country LIKE ‘Z%’

    From statistics, we know approximately how many countries have a name beginning with Z. SQL Server also has a histogram for the NationId column in the Customers table. If we had specified the list of NationId values (with equality on to both tables), SQL Server could use the more detailed information from the histogram to make a row estimate. But because we specified the Country name column that is only in the Nation table, we must use the average number of customers per county (in the density vector) multiplied by the number of countries to estimate rows after the join to customers.

    And next of course, all customers are not alike. There are customers who place a small, medium or large number of orders. So why expend a great deal of effort to find a plan from all the possible join orders and index combination based on only estimates of rows when it is known that data distribution in each succeeding table is heavily skewed. Why not pre-execute the tables for which a SARG was specified to get the column values used in the join to next table so that the more detailed histogram distribution information. This technique could be pushed to multiple levels depending on the initially assessed plan cost, and perhaps controlled by a query hint.

    For example, in the above query, suppose NationId’s 19, 37, 42, and 59 meet the criteria Country beginning with Z. The optimizer would next look at the Customers table histogram on NationId for these values to estimate rows after the join. If the situation warrants, the next level could be examined as well.

    It could be argued that the query optimizer should not execute the query to determine the plan, but why follow that principle if the cost of query optimization is excessively high (several seconds) in relation to relatively minor effort to make a more extensive reconnaissance (of tens or hundreds of milli-seconds)? Especially considering that the reliability of row estimates becomes progressively worse after each join or other operation beyond the original source?

    This technique should probably be used when there are tables with search arguments joining to tables on columns with highly skewed distribution. The first implementation might be activated only be a query hint until some maturity is achieved, followed by greater use.

    Presumably there might be a cost threshold as well. I would prefer not to tie it with parallelism. Of course, given the nature of modern systems, it really is time for the cost threshold for parallelism and max degree of parallelism to have graduated controls, instead of the single setting on-off.

    Side note 1

    OK, now forget what I said at the beginning and I will gripe about SQL Server default statistics. It has been discussed else where that SQL Server uses random page samples and not random row samples, as this is a much less expensive way to collect data. It does use an index for which the column is not a lead key if available, to improve randomness. Still, I have notice a severe sensitivity to sampling percentage in cases where the column value is correlated with (page) storage location.

    So I suggest that as the page sampling is in progress, a count of new values found in each successive page sampled versus existing values be kept. If the number of new values found falls off sharply, then most distinct values have probably been found, and the interpretation of the existing sample is that its distribution should be scaled by ratio of total rows to the rows sampled. If almost of the values in the last few pages sampled are new (not previously found), then the interpretation should be that the number of distinct values should be scaled by the total to sample ratio. And some blend when there is an intermediate number of new values versus previously found values in each successive page.

    Addendum

    The query optimization/plan compile process is single threaded. The modern microprocessor might be 3GHz, so a 10 sec compile is 30 Billion cpu-cycles. And I have seen compiles run more than 100 sec? One query even broke SQL Server, of course that was a set of deeply nested, repeating CTE's that should have been PIVOT/UNPIVOT. So why the principle of optimizing based on unreliable estimates when an index seek is a mere 10 micro-sec? Presumably the key column statistics have already been decoded.

    It would be nice to have a more powerful query optimizer, but there is a method for writing SQL to get a specific execution plan, Bushy Joins. Of course the other element in this is known what the correct execution plan is. This involves not what the query optimizer uses for cost model, but what the true cost model is.

  • TPC-H and Columnstore (Update)

    Earlier I had commented on the TPC-H results published in April of this year for SQL Server 2014 using clustered column store storage, noting that two of the 22 TPC-H queries did not perform well in column store. I had speculated on the reason without investigation (I should have learned by now not to do this), that perhaps the cause was that the row store result benefited from date correlation optimization. Thomas suggested otherwise (see below) pointing out that column store has an alternative mechanism of greater general usefulness in the keeping min/max on each columns, along with citing the join to Customers as a more likely explanation, evident in the query plan (which is why one should always provide the plan).

    Thomas Kejser Comments
    I am not sure your theory is correct in the case of Q10. It is noteworthy that the column store requires that the join with CUSTOMER is performance before the sort on revenue. The row store on the other hand can do a loop join (so can the column store, but that is not the plan you get it seem).

    This must mean that the sort buffer is significantly larger for the column store (as reflected in the plan estimates) - which in turn can cause a rather significant memory consumption. It is also noteworthy that the column store does not seem to push the return flag predicate into the storage engine.

    With column storage segments storing the min/max of all values contain in each column, it is unclear if the date correlation provides any benefit that isn't already gain from the segment header.

    Another odd thing about the column store plan of Q10 is that the join of LINEITEM/ORDER is hashed, while the probe happens on CUSTOMER. Unless the predicate on RETURNFLAG is very selective (I don't recall) this is the wrong way around and may cause further spilling

    This is easy enough to test. First strategy is to remove the CUSTOMER and NATION tables from the query, making it a pure Below is the test version of Query 10.

    SELECT TOP 20 O_CUSTKEY, SUM(L_EXTENDEDPRICE*(1-L_DISCOUNT)) AS REVENUE
    FROM ORDERS, LINEITEM
    WHERE L_ORDERKEY = O_ORDERKEY AND O_ORDERDATE>= '1993-10-01' AND
    O_ORDERDATE < dateadd(mm, 3, cast('1993-10-01' as date)) AND
    L_RETURNFLAG = 'R'
    GROUP BY O_CUSTKEY
    ORDER BY REVENUE DESC

    The execution plan for the test query is below with row-store. (Top operation to the left not shown for compactness).

    tpcE

    The execution plan for the test query is below with column-store. (Top operation also not shown).

    tpcE

    The two plans are essentially the same, with the difference being that the column-store plans applies the ReturnFlag as a filter operation instead of as a predicate in the LINEITEM access. I suppose this is because each column is stored separately, or perhaps this is just the way the column-store plan is shown. About 25% of rows in LINEITEM meet the ReturnFlag = R condition.

    On my 1 socket 4 core, HT enabled test system, at Scale Factor 10 (SF10), the SQL Server query execution statistics for the original version of Q10 are:
    Row Store CPU time =  5876 ms, elapsed time =  814ms
    Col Store  CPU time = 10826 ms, elapsed time = 1758ms

    This is somewhat in-line with the 3 official TPC-H reports at SF 1000, 3000 and 10000 (1, 3 and 10TB) compared against different systems and SQL Server 2012 or earlier.

    For just the core ORDER - LINEITEM query
    Row Store CPU time = 5248 ms, elapsed time = 769ms
    Col Store   CPU time = 4030 ms, elapsed time = 565ms

    So it is clear that TK had the correct explanation for the poor column store performance relative to row store in the case of Q10. The counter test for my original suggestion, is to explicitly apply the date range on LINEITEM discovered by the date correlation optimization in row-store, L_SHIPDATE between 1993-09-20 and 1994-06-17. As pointed out earlier, the actual ship date range 1993-10-02 and 1994-05-02. This further improved the Columnstore result to
    Col Store   CPU time = 2686 ms, elapsed time = 403ms

    This is a small improvement over the existing Columnstore min/max feature. My thinking is that the row store date correlation feature is not particularly useful in real world databases with highly irregular date correlation, and that if such date correlation did exist, the analyst should spell it out rather than depend on a database engine feature. I am tempted to speculate that it might be better to partition on join columns instead of date range, but perhaps I should not do so without investigation? unless of course, this prompts someone else to do the investigation.

    Now that we know were the problem occurred in Q10, we can attempt to rewrite the query to avoid the error, as shown below.

    SELECT C_CUSTKEY, C_NAME, REVENUE, C_ACCTBAL, N_NAME, C_ADDRESS, C_PHONE, C_COMMENT
    FROM (
     SELECT TOP 20 O_CUSTKEY, SUM(L_EXTENDEDPRICE*(1-L_DISCOUNT)) AS REVENUE
     FROM CUSTOMER, ORDERS, LINEITEM, NATION
     WHERE L_ORDERKEY = O_ORDERKEY AND O_ORDERDATE>= '1993-10-01' AND
     O_ORDERDATE < dateadd(mm, 3, cast('1993-10-01' as date)) AND
     L_RETURNFLAG = 'R'
     GROUP BY O_CUSTKEY
     ORDER BY REVENUE DESC
    ) x
    JOIN CUSTOMER ON C_CUSTKEY = O_CUSTKEY
    JOIN NATION ON C_NATIONKEY = N_NATIONKEY
    ORDER BY REVENUE DESC

    The alternate query improved both the row and column-store query plans in pushing out the join to Customers and Nation to after the Top clause. The row-store plans is:

    tpcE tpcE

    The column-store plan is below.

    tpcE tpcE

    The impact is minimal in the row-store plan because the reduced number of index seeks for Customers from 115 to 20 is small in the overall query. For the column store plan, the performance retains most of the gains achieved in the Order-Lineitem only test.
    Col Store   CPU time = 4218 ms, elapsed time = 598ms

    In my test system, I have Q4 as 3 times faster with column-store over row-store, so I do not know why the published reports have it as comparable or slower.

    ps
    columnstore is a very powerful. but the query optimizer is as mature as row store. So pay attention to the query plan.
    this is an update to my previous post on this topic, not about updateable columnstore, which is updateable.

  • TPC-H Benchmarks on SQL Server 2014 with Columnstore

    Three TPC-H benchmark results were published in April of this year at SQL Server 2014 launch, where the new updateable columnstore feature was used. SQL Server 2012 had non-updateable columnstore that required the base table to exist in rowstore form. This was not used in the one published TPC-H benchmark result on SQL Server 2012, which includes two refresh stored procedures, one inserting rows, the second deleting rows. It is possible that the TPC-H rules do not allow a view to union two tables? and perhaps a delete via the partitioning feature? (meaning the delete range must match the partition boundaries). Another possibility is that SQL Server 2012 columnstore was considered to be a multi-column index which is also prohibited to reflect the principle of being ad-hoc queries.

     

    SQL Server Columnstore

    First a few quick words on SQL Server columnstore. Columnstore is not actually an index of the b-tree index form. The MSDN Columnstore Indexes Described states that columnstore index is "a technology for storing, retrieving and managing data by using a columnar data format, called a columnstore." In SQL Server 2012, it is called a nonclustered columnstore index not because it is nonclustered or an index, but because the base table must exist in traditional rowstore form. In SQL Server 2014, there is a clustered columnstore index not because data is stored in order of the index key, as there is no key, but rather that there is no rowstore table, just the columnstore.

     

    TPC-H details Date Columns and Query SARGs

    The full details of the TPC-H Decision Support benchmark are described on the TPC.org website. There are a few details of relevance to the use columnstore. The largest table in TPC-H is LINEITEM, which has 3 date columns, ShipDate, ReceiptDate and CommitDate, and is clustered on ShipDate. The second largest table is ORDERS, clustered on the one date column OrderDate. These two tables are joined on OrderKey. There is correlation between values in the date columns in these two tables, some natural, and others based on reasonable business conditions. ShipDate must be greater than OrderDate obviously, and is also no more than 121 days greater than OrderDate per benchmark specification built into the data generator. CommitDate is between -89 and 91 days of ShipDate. ReceiptDate is between 1 to 30 days after ShipDate. The date values ranges from Jan 1992 to Dec 1998.

    There are 22 Select queries in the TPC-H benchmark, along with the 2 refresh stored procedures. Many of the Select queries specify a date range on one of the date columns or a lower bound one date column and an upper bound on a different column. Ideally, for queries that target rows in a limited date range, we would like to have upper and lower bounds on for the cluster keys on both the ORDERS and LINEITEM tables, OrderDate and ShipDate. However the TPC-H rules do not permit re-writing the query SARGs based on inferable knowledge.

    That said, apparently the rules do not preclude the query optimizer from discovering such knowledge. One of the other RDBMSs was probably first to do this, and Microsoft followed suit in order to be competitive in the TPC-H benchmark with the Date Correlation Optimization feature in 2005. Personally, I am not aware of any production server using this feature. Realistically, any organization that was having query performance issues related to date range bounds would probably have directed the analyst to apply appropriate date bounds on the cluster key. This is most probably a benchmark specific optimization feature.

    The date correlation optimization statistics do not exist when using clustered columnstore, because there is no underlying rowstore table with foreign key relations? The date correlation statistics do exist when using rowstore tables with foreign keys and are used by nonclustered columnstore indexes?

     

    TPC-H on SQL Server 2014 with Columnstore

    That said, let us now look at the 3 new SQL Server 2014 TPC-H results published making use of the new clustered columnstore indexes. One is from IBM at Scale Factor 1000 (1TB) and two from HP at 3TB and 10TB respectively. The new results are compared to prior results with traditional rowstore on previous versions of SQL Server and previous generation server systems.

    Because Columnstore is not really an index in the b-tree sense, given that queries frequently involve date ranges, it is presumed to be important to use partitioning with Columnstore. The three new TPC-H reports on SQL Server 2014 partition both ORDERS and LINEITEM by OrderDate and ShipDate respectively (the cluster key in previous versions) with a partition interval of 1 week (7 years x 52 weeks per year = 364 partitions). Perhaps of interest, the scripts show that a rowstore partitioned index is first build before building the partitioned clustered columnstore index.

     

    TPC-H at SF 1000 (1TB)

    The new TPC-H 1TB result on SQL Server 2014 using columnstore is compared with 3 previous results on SQL Server 2008 R2. There is a difference in memory configurations between the four systems below. For SQL Server 2014 with clustered columnstore indexes, the TPC-H SF1000 total database size is just under 430GB, so with 1TB memory, the benchmark is running entirely in memory after the initial data load, with the exception of hash operation spills to tempdb.

    Vendor
    System
    Processor
    GHz
    Sock
    ets
    Cores
    Threads
    MemSQLQphHPowerThrough
    put
    Date
    IBM
    x3850 X6
    E7-4890 v2
    2.8
    460/12015362014519.976695,445388,7794/16/14
    HP
    DL980 G7
    E7-4870
    2.4
    880/16020482008R2219,888233,119207,4078/30/11
    IBM
    x3850 X5
    E7-8870
    2.4
    880/8020482008R2173,961200,889150,6355/20/11
    Cisco
    UCS C460
    E7-4870
    2.4
    440/?10242008R2134,117156,157115,18812/7/11

    For rowstore, the TPC-H SF1000 total database size is 1420GB, so the two systems with 2TB memory are mostly running with data in memory, again except for the initial load and spills to tempdb. There is definitely disk IO for data in the Cisco system at 1TB physical memory. The performance impact is noticeable, but probably not as severe as one might think based on how people talk of memory. The reason is that all of these systems make correct use of massively parallel IO channels to storage capable of 10GB/s plus table scans, and many of these also use SSD storage capable of more random IOPS than SQL Server can consume even at very high parallelism.

    The new SQL Server 2014 result is 2.36X higher on composite score (QphH) and 3X higher in the Power test than previous versions with conventional rowstore.

    The 22 individual query run times from the Power test at 1TB are shown below. The SQL Server 2014 result is the left item in the legend label 4x60 (sockets/cores) and for the succeeding charts as well.

    tpcE

    Query 1, a single table aggregation, is more than 10 times faster on SQL Server 2014 using columnstore on 60 cores (Ivy-Bridge, 2.8GHz) than 2008R2 using rowstore, on 80 cores(Westmere-EX). Per TPC-H benchmark procedure, the test is run immediately after data load and index creation? The second largest speed-up is Query 16, joining 3 tables, at 6.4X.

    Query 10 is 40% slower with columnstore, and Query 4 about the same between columnstore and conventional. This query is listed near the end of this section on TPC-H.

    Notice that in the 3 SQL Server 2008R2 results, Query 2 becomes slower as the degree of parallelism increases from 40 cores (threads unspecified) to 80 cores/80 threads and then to 80 cores/160 threads. Elsewhere I had commented that SQL Server really needs a graduated approach to parallelism instead of the all or nothing approach.

    TPC-H at SF 3000 (3TB)

    The new TPC-H 3TB result on SQL Server 2014 is compared with a previous result on SQL Server 2008R2. Here, the difference in memory is a significant contributor. The SQL Server 2014 system has more memory than the columnstore database, while 2008R2 systems has much less memory than the 3TB rowstore database (4.5TB).

    Vendor
    System
    Processor
    GHz
    Sock
    ets
    Cores
    Threads
    MemSQLQphHPowerThroughputDate
    HP
    DL580 G8
    E7-4890 v2
    2.8
    460
    120
    30722014461,837631,309337,8594/15/14
    HP
    DL980 G7
    X7560
    2.27
    864
    128
    5122008R2162,601185,297142,6856/21/10

    I am supposing that the reason the HP 2010 report only configured 512GB (128 x 4GB priced $29,440 in 2010) was that there would not be a significant performance improvement for the TPC-H 3TB result at either 1TB or even 2TB memory in relation to the higher price (128 x 16GB priced $115K in 2011).

    Several of the TPC-H queries involve nearly full table scans of the large tables. If there is not sufficient memory for the entire database, then the next objective is to have sufficient memory for reducing the spill to disk in hash operations? HP may have elected for the better price-performance? Or perhaps someone just wanted to make a point. The point being that it is important for the SQL Server engine to function correctly when heavy IO is required.

    In the SQL Server 2014 result, the system has 3TB memory (96 x 32GB priced $96K in 2014) which is sufficient to hold the entire data set for TPC-H 3TB in columnstore.

    The overall composite score is 2.8X higher with columnstore and 3.4X higher on the Power test.

    The 22 individual query run times from the Power test at 3TB are shown below.

    tpcE

    The largest gain with column-store is Query 19 at 19.7X. Query 4 and 10 show degradation, similar to the case at SF1000.

    TPC-H at SF 10000 (10TB)

    The new TPC-H 10TB result on SQL Server 2014 is compared with a previous result on SQL Server 2012. Strangely, supporting documentation for the HP 2013 report on SQL Server 2012 is missing so there is no indication as to whether nonclustered columnstore is used? I am guessing that columnstore was not used because the results are in line with expectations on rowstore.

    Vendor
    System
    Processor
    GHz
    Sock
    ets
    Cores
    Threads
    MemSQLQphHPowerThroughputDate
    HP
    DL580 G8
    E7-4890 v2
    2.8
    460
    120
    3072 2014404,005631,309337,8594/15/14
    HP
    DL980 G7
    E7-4870
    2.4
    880
    160
    4096 2012158,108185,297142,6856/21/11

    The full data size in columnstore at SF 10000 should be 5TB, and 14TB in rowstore, so there should have been heavy disk IO in both results.

    The overall composite score is 2.55X higher with column-store and 3.1X higher on the Power test.

    The 22 individual query run times from the Power test at 10TB are shown below.

    tpcE

    The largest gain with column-store is Query 6 at 23.2X. Query 4 and 10 show degradation as in the two previous cases.

     

    TPC-H Query 10

    Below is Query 10. This query is consistently slower in columnstore relative to rowstore.

     

    /* TPC_H Query 10 - Returned Item Reporting */

    SELECT TOP 20 C_CUSTKEY, C_NAME, SUM(L_EXTENDEDPRICE*(1-L_DISCOUNT)) AS REVENUE, C_ACCTBAL,
    N_NAME, C_ADDRESS, C_PHONE, C_COMMENT
    FROM CUSTOMER, ORDERS, LINEITEM, NATION
    WHERE C_CUSTKEY = O_CUSTKEY AND L_ORDERKEY = O_ORDERKEY AND O_ORDERDATE>= '1993-10-01' AND
    O_ORDERDATE < dateadd(mm, 3, cast('1993-10-01' as date)) AND
    L_RETURNFLAG = 'R' AND C_NATIONKEY = N_NATIONKEY
    GROUP BY C_CUSTKEY, C_NAME, C_ACCTBAL, C_PHONE, N_NAME, C_ADDRESS, C_COMMENT
    ORDER BY REVENUE DESC

    The execution plan for rowstore at SF 10 is shown below.

    tpcE
    (Click for full-size)

    The execution plan for columnstore at SF 10 is shown below.

    tpcE
    (Click for full-size)

    Below are the details on ORDERS and LINEITEM from the rowstore plan.

    tpcE tpcE

    Notice that there are seek predicates on LINEITEM for 1993-09-20 to 1994-06-17. The actual range should be 1993-10-02 to 1994-05-02, for 1 day after the OrderDate lower bound and 121 days after the OrderDate upper bound.

    Below are the details on ORDERS and LINEITEM from the columnstore plan.

    tpcE tpcE

    In columnstore, every operation is a scan. There is a predicate for the ORDERS table but not on the LINEITEM table. Presumably storage engine must scan entire set of LINEITEM partitions while only scanning the ORDERS partitions encompassing the SARG date range

    I am thinking the reason is that with date correlation in conventional row-storage, the SQL Server query optimizer knows that the data range in LINEITEM ShipDate is also restricted by the lower OrderDate and the upper OrderDate plus 121 days, corresponding to 1 day after the lower bound on OrderDate to 121 days after the upper bound on OrderDate.

    TPC-H Query 4

    TPC-H Query 4 below is slower than row storage in the 3 and 10TB results. I am thinking that the reason is the same?

     

    /* TPC_H Query 4 - Order Priority Checking */

    SELECT O_ORDERPRIORITY, COUNT(*) AS ORDER_COUNT FROM ORDERS
    WHERE O_ORDERDATE >= '1993-07-01' AND O_ORDERDATE < dateadd(mm,3, cast('1993-07-01' as date))
    AND EXISTS (SELECT * FROM LINEITEM WHERE L_ORDERKEY = O_ORDERKEY AND L_COMMITDATE < L_RECEIPTDATE)
    GROUP BY O_ORDERPRIORITY

     

    See TPCH Query Plans for the TPC-H reference queries and execution plans at SF1 on SQL Server 2005. The parent page TPCH Interim has links for the SF1000 query plans with and without parallelism.

    TPC-H Columnstore Summary

    As with every other new feature, Columnstore is a really interesting new technology. But think hard about what is really happening, experiment, and remember to get good execution statistics and plans prior to making changes, then get the new execution statistics and plans after the change.

    One reason I like to look at official TPC-H benchmark results over "real-world" is that the benchmark system is properly configured for both before and after results. There is a significant difference in the data size involved for each query between rowstore and columnstore. If the reference system has a poor storage system (and how often have we seen this? this is guaranteed when the SAN vendor assisted in configuration), then it is possible to produce almost any performance ratio.

     

    TPC-E

    The charts below show the progression of performance over time for the selected TPC-E results spanning Core 2, Nehalem, Sandy Bridge and Ivy processors at 2, 4 and 8 sockets.

    tpcE

    For the 2-socket systems, West-1 is from the first set of TPC-E results reported for Westmere X5680 with HDD storage and West-2 is the later X5690 report with SSD storage. Both are 6-core Westmere-EP processors. The West-3 is the E7-2870 10-core (Westmere-EX) on SSD storage.

    For the 4-socket systems, West-1 is on HDD storage, and West-2 on SSD, both 2K8R2 and 1TB memory. The West-3 is on Win/SQL 2012, 2TB memory and SSD storage.

    The same data is shown below with reverse organization showing scaling with sockets for each of the processor architectures.

    tpcE

    Notes: Nehalem
    2-socket 4-core, 2.93GHz (11.72 core x GHz), 4 & 8-socket 8-core 2.26GHz (18.08 core x GHz)
    Westmere
    2-socket 6-core, 3.46GHz (20.76 core x GHz/socket), 4 & 8-socket is 10-core 2.4GHz (24 core x GHz)
    Sandy Bridge
    2-socket 8-core, 2.9GHz (23.2 core x GHz/socket), 4-socket is 8-core 2.7GHz (21.6 core x GHz)
    Ivy Bridge
    2-socket 12-core, 2.7GHz (32.4 core x GHz/socket), 4 & 8-socket is 15-core 2.8GHz (42 core x GHz)

  • Detecting Hyper-Threading state

    To interpret performance counters and execution statistics correctly, it is necessary to know state of Hyper-Threading (on or off). In principle, at low overall CPU utilization, for non-parallel execution plans, it should not matter whether HT is enabled or not. Of course, DBA life is never that simple (see my other blogs on HT). The state of HT does matter at high overall utilization and in parallel execution plans depending on the DOP. SQL Server does seem to try to allocate threads on distinct physical cores at intermediate DOP (DOP less than or equal to the number of physical cores).

    Suppose for example, that maximum throughput on 10 physical cores is 10000 call/s with HT off and 14000 with HT on (overall CPU near 100%). Then the average CPU (worker time) per call is 1ms with HT off, and 1.43 ms with HT on as there are twice as much available worker time with HT on.

    In a very well tuned OLTP system, we might have very steady average CPU per call as call volume increases from low overall CPU utilization to near saturation with HT off. With HT on, at low overall CPU, the average CPU per call is the same as with HT off, but average CPU per call increases at some point when there is sharing of physical cores between concurrently running queries. But this is still good because system wide throughput capability has increased.

    In a not well tuned database application, there could be contention between concurrent queries that causes average CPU per call to increase as overall system CPU load increases. Without knowing the state of HT, it is hard to make the assessment as to which situation has occurred.

    If we have direct sysadmin access to the OS, we could make calls (via WMI) to determine the processor model number, the total number of sockets, the total number of logical processors, then determine with a lookup table matching processor model to the number of physical cores. (again, a pain)
    (per LondonDBA, WMI Win32_Processor does report the number of sockets, physical cores, and logical, I must have been thinking of an older API that was not HT aware or even multi-core aware)
    But we do not always have sysadmin access to the host OS, as many organization believe separation of DBA, infrastructure (not to mention storage) is a good thing, and even better when these groups do not communicate with each other, (let alone) working together with a common mission.

    SQL Server version 2005 was helpful in the DMV sys.dm_os_sys_info
    which had two columns: cpu_count and hyperthread_ratio
    defined in version 2005 as:
    1) "Number of logical CPUs on the system."
    2) "Ratio of the number of logical and physical processors."
    respectively.

    In version 2008, RTM and R2, the definition of hyperthread_ratio was changed to:
    "Ratio of the number of logical or physical cores that are exposed by one physical processor package."

    in 2012 and 2014, slightly different wording but same meaning:
    "Specifies the ratio of the number of logical or physical cores that are exposed by one physical processor package."

    WTF???
    why the change in definition?
    There are actually 3 pieces of information we are interested in:
    a) the number of sockets
    b) the number of physical cores per socket
    c) the state of HT (or the logical processors per socket)

    In 2005, we have information to determine the product of A and B, and the value of C, but not atomic values of A and B,
    In the 2008 and later version, we can determine A (using the ratio) and the composite product of A x B x C, but not atomic values of B and C.

    Ok, then I noticed that in SQL Server version 2012 and 14, there would be a line in the log of the form:
    "SQL Server detected 4 sockets with 10 cores per socket and 10 logical processors per socket, 40 total logical processors; using 40 logical processors based on SQL Server licensing."

    So from this, I have A, B and C, even though I must parse the error log for this info. In version 2008R, the info is only: "Detected 40 CPUs."
    which has no additional information to what is in the DMV. It would be helpful if this information were available directly from the DMV, but then life might be easier?

    Addendum

    EXEC sys.xp_readerrorlog 0, 1, "detected", "socket"

  • Top Clause and Other Factors in Problematic Execution Plans

    Three years ago, I conducted an extensive investigation on a SQL Server system running kCura's Relativity document e-discovery application. It was fascinating to see the broad range of problematic queries all from one application. This provided good material for my presentation Modern Performance which focuses on the more spectacular problems that can occur with a cost based query optimizer.

    (I am not sure why the images are not showing up. See the alternate link Top Clause)

    It should be pointed that unlike a typical application in which the key queries can be rigorously tuned, Relativity must generate the SQL from options in the UI. The objective of Relativity is to support complicated searches, with heavy emphasis on queries of the form: IN set A but NOT IN set B. This causes problems in row estimation because there is no generally valid logic to assess whether there is overlap between Set A and Set B even if row estimates on the individual sets are possible.

    If this were not difficult enough, there can also be nested AND/OR combinations. SQL Server has difficulty generating good execution plans even for a single AND/OR combination. It is unfortunate that the most direct conversion of natural (user-oriented) logic to a SQL expression (that Relativity employs) just happens to be the form for which the SQL Server query optimizer has difficulties in producing a good execution plans.

    Relativity 8.1

    This article is based on a brief observation of Relativity version 8.1. The first Relativity article was originally based on version 7.3-7.5, and later updated with observations of version 8.0.

    Relativity Architecture

    The architecture of Relativity first employs a SELECT COUNT query to determine the number of rows that meet a specific search query. A second call is then made in the form of a SELECT TOP 1000 query to retrieve just the document identity column (ArtifactID) values. There should be a third query that retrieves all the desired columns for specific ArtifactID values, but this query is very low cost and does not warrant discussion here. If more than 1000 rows are needed, then the next query will be TOP 6000.

    One observation is that even when the COUNT query indicates that there are fewer than 1000 rows, the ArtifactID query is still issued with the TOP 1000 clause. In earlier versions of Relativity (7.x), the TOP 1000 query was issued even when the COUNT query indicated zero rows. Perhaps an oversight by the developers in not realizing zero rows are returned when COUNT is zero.

    One aspect of Relativity architecture that is correct is the not enabling plan reuse. The expectation is that individual searches probably have high CPU for execution than for compile and that each parameter set could have skewed distribution. However, in issuing explicit SQL, there is also no opportunity to fix problematic SQL. Hence it is important the Relativity anticipates as many problems as possible and employ good strategies for know issues. We could wish for such, and we would still be wishing.

    Problematic SQL in Relativity 8.1

    There are several difficulties in Relativity queries, three of which are covered here. One is due to the SELECT COUNT followed by TOP 1000 architecture. The second is due to the use of the direct translation of natural (user) logic to SQL rather than a form of SQL for which the SQL Server query optimizer happens to produce very good execution plans. The third is in parallelism strategy. The data in Relativity is expected to have heavily skewed distribution so is necessary to watch for situations where this produces ineffective parallel execution plans.

    Relativity Query Example

    Below is the COUNT form of an example search query to be studied in greater detail.

    SELECT COUNT( [Document].[ArtifactID])
    FROM eddsdbo.[Document] WITH(NOLOCK)
    LEFT JOIN eddsdbo.[Custodian] AS [o1000022_f2154932] (NOLOCK)
    ON [o1000022_f2154932].[ArtifactID] = [Document].[Custodian]
    WHERE [Document].[AccessControlListID_D] IN (1,1000062,1000063,...)
    AND
    (
      [o1000022_f2154932].[ArtifactID] IN (22552945, 22552935, 22552925, 22552915, 22552905, 14768277,...)
     OR (EXISTS(
      SELECT [f2154930f2154931].[f2154931ArtifactID]
      FROM eddsdbo.[f2154930f2154931] (NOLOCK)
      LEFT JOIN eddsdbo.[Custodian] (NOLOCK)
      ON [Custodian].[ArtifactID] = [f2154930f2154931].[f2154931ArtifactID] 
      WHERE [f2154930f2154931].[f2154930ArtifactID] = [Document].[ArtifactID]
      AND [Custodian].[ArtifactID] IS NOT NULL
      AND ([Custodian].[ArtifactID] IN (22552945, 22552935, 22552925, 22552915, 22552905, 14768277,...))
     ))
    )

    The actual number of values in the AccessControlListID_D IN clause is 113. The actual number of values in each of the Custodian.ArtifactID IN clauses is somewhat over 100, with both sets being identical.

    Below is subsequent TOP 1000 query that is issued after the COUNT query. In the 7.x versions, the TOP 1000 clause is present regardless of the number of rows indicated by the COUNT query, and is in fact issued even if COUNT reports zero rows. It is also very possible that either one of the COUNT and TOP queries or both are very expensive regardless of the actual number of rows.

    SELECT TOP 1000 [Document].[ArtifactID]
    FROM eddsdbo.[Document] WITH(NOLOCK)
    LEFT JOIN eddsdbo.[Custodian] AS [o1000022_f2154932] (NOLOCK)
    ON [o1000022_f2154932].[ArtifactID] = [Document].[Custodian]
    WHERE [Document].[AccessControlListID_D] IN (1,1000062,1000063,...)
    AND
    (
      [o1000022_f2154932].[ArtifactID] IN (22552945,22552935, 22552925, 22552915, 22552905, 14768277,..)
     OR (EXISTS(
      SELECT [f2154930f2154931].[f2154931ArtifactID]
      FROM eddsdbo.[f2154930f2154931] (NOLOCK)
      LEFT JOIN eddsdbo.[Custodian] (NOLOCK)
      ON [Custodian].[ArtifactID] = [f2154930f2154931].[f2154931ArtifactID] 
      WHERE [f2154930f2154931].[f2154930ArtifactID] = [Document].[ArtifactID]
      AND  [Custodian].[ArtifactID] IS NOT NULL
      AND ([Custodian].[ArtifactID] IN (22552945, 22552935, 22552925, 22552915, 22552905, 14768277,...))
     ))
    )
    ORDER BY [Document].[ArtifactID]

    Alternative Queries

    For the above search with the actual data set, there are in fact 232 rows. Three alternatives queries are examined here. One is the above query without the TOP clause. The second alternative is the above query (including the TOP clause) but with an index hint applied. The third alternative tested is the form below, with the OR clause replaced by a UNION.

    SELECT [Document].[ArtifactID]
    FROM eddsdbo.[Document] WITH(NOLOCK)
    LEFT JOIN eddsdbo.[Custodian] AS [o1000022_f2154932] (NOLOCK)
    ON [o1000022_f2154932].[ArtifactID] = [Document].[Custodian]
    WHERE [Document].[AccessControlListID_D] IN (1,1000062,1000063)
    AND
    (
      [o1000022_f2154932].[ArtifactID] IN (22552945, 22552935, 22552925, 22552915, 22552905, 14768277,...)
    )
    UNION
    SELECT  [Document].[ArtifactID]
    FROM eddsdbo.[Document] WITH(NOLOCK)
    LEFT JOIN eddsdbo.[Custodian] AS [o1000022_f2154932] (NOLOCK)
    ON [o1000022_f2154932].[ArtifactID] = [Document].[Custodian]
    WHERE [Document].[AccessControlListID_D] IN (1,1000062,1000063,...)
    AND
    (
     (EXISTS(
      SELECT [f2154930f2154931].[f2154931ArtifactID]
      FROM eddsdbo.[f2154930f2154931] (NOLOCK)
      LEFT JOIN eddsdbo.[Custodian] (NOLOCK)
      ON [Custodian].[ArtifactID] = [f2154930f2154931].[f2154931ArtifactID] 
      WHERE [f2154930f2154931].[f2154930ArtifactID] = [Document].[ArtifactID]
      AND  [Custodian].[ArtifactID] IS NOT NULL
      AND ([Custodian].[ArtifactID] IN (22552945, 22552935, 22552925, 22552915, 22552905, 14768277,...))
     ))
    )
    ORDER BY [Document].[ArtifactID]

    For this particular case, the UNION will produce exactly the same set of rows as the original query because only the Document table ArtifactID column is in the select list and this column is the primary key of the Document table. In the more general case with columns from more than one table in the SELECT list, the UNION form could have a different row set than the original OR form. The architecture of Relativity is such that the OR forms can be converted to UNION while maintaining correct results.

    Earlier, it was mentioned that the SQL Server query optimizer can have problems in producing a good execution plan for queries with a combination of AND and OR clauses. The form shown above is set A OR set B, with A having an AND clause. The other form of this is set A AND set B, with either A or B having an OR clause. The second form could probably be handled by converting the AND to an INNER JOIN, which would require both conditions to be true, but a full study has not been done on this.

    Relativity Query Execution Plans

    Now consider the estimated execution plans for the 5 queries: 1) COUNT, 2) TOP, 3) no TOP, 4) index hint, and 5) UNION.

    The SELECT COUNT estimated execution plan is shown below (full plan).

    CountEst

    The SELECT TOP 1000 estimated execution plan (full plan).

    Note top right operation in the above execution plan for the TOP 1000 query. The operation is an Index Scan on index EV_ArtifactID. The index lead key is ArtifactID, and has all the columns required for this query.

    Below left is the Index Scan detail. Notice that the Estimated I/O cost is 85, corresponding to approximately 892MB (I/O cost 1 = 1350 pages or 10.5MB). Below right is the COUNT query Nested Loops operation at the top left of the plan.

    Below are the left most operator showing total plan cost for the COUNT and TOP 1000 queries respectively.

    ,

    The TOP 1000 plan cost is 0.31269 even though the Index Scan on Document table index EV_ArtifactID has an I/O of 85.38. This is because the execution plan is expecting the first 1111 rows from Documents to be sufficient to meet the total required 1000 rows. The plan cost for the COUNT query is much higher at 519 as all rows must be evaluated to test the match to the search arguments. The Document Index Scan in the COUNT query uses the IX_AccessControlListID_D index because this index is narrow with cost of 35.46 (272MB)

    Below is the estimated plan for the SELECT query without the TOP clause (full plan)

    ,

    The estimated row count is 11,569,800, exactly the same as in the COUNT query at the Nested Loops operation just before the rows are aggregated into a COUNT value.

    There are 12.8M rows in the Document table, so the query optimizer believes that about 90% of row meet the search argument (with OR clause). The execution plan is the same as the COUNT query except that the return list requires a sort operation contributing to the higher overall plan cost of 720.

    Below is the estimated plan for the TOP query but with an index hint applied (full plan)

    The use of the index hint results in the query optimizer not changing the join order. Hence when using hints, it is also necessary to write the query in the form with the desired join order. The hinted index lead key does not match in the ORDER BY clause, so the entire index is scanned and then sorted. The plan cost 197 is more than the original TOP query, but less than the no TOP query. The reason is that the query optimizer believes it can exit the execution plan once the TOP 1000 rows have been found.

    Below is the estimated execution plan for the UNION query. (full plan)



    The Estimated Number of Rows is 3.2M and the plan cost is 92.8.

    Curiously, the number of rows from each of the two sub-queries is much less at 23,764 and 923,036.

    One would think that the UNION of the two result sets should at least equal to the larger and could be as high as the sum.

    Relativity Query Execution Times

    Below are the execution time in milli-seconds for both CPU and elapsed times of each of the 5 queries. The first is the COUNT query, followed by the Relativity standard TOP 1000 query. Next are the 3 variations considered: without the TOP clause, with an Index Hint and finally a UNION structure instead of OR in the WHERE clause.

    QueryCPU timeelapsed time
    COUNT13827969109
    TOP 1000152148152298
    no TOP14397371857
    index hint16767122016
    UNION1665324

    The standard Relativity TOP query does not have a parallel execution plan as the plan cost (0.31269) is below the cost threshold for parallelism. Even though some of the operations in the plan show a cost greater than 0.3, the Top clause believes that the query can terminate after few rows from Documents (estimate 1111) have been examined.

    In this example, the WHERE clause arguments are highly selective, and fewer than 1000 rows meet the conditions. So what happens is that the full set of 12.8M rows from the right most loop join outer source must be processed, as the Top clause "exit" criteria is never reached. Furthermore, the execution in not parallel, because it is believed to be a low cost plan.

    Note the fat arrows in the SELECT TOP 1000 actual execution plan (full plan) below.

    There is not a significant difference in CPU times between the COUNT, regular TOP, no TOP, and TOP + index hint query plans. The differences are mostly in the elapsed time. The regular TOP query has the longest elapsed time being a non-parallel plan as explained above. The COUNT and no TOP queries have elapsed times approximately one-half of the CPU time, corresponding to a 2X speed-up with parallelism even though the actual degree of parallelism was 8 (on separate physical cores).

    Parallel Execution Plans

    There are two critical factors/challenges in parallel execution plans. One is to divide the work evenly among multiple threads (each running on different cores). The second is to have sufficiently large granularity of work on each thread before some form of synchronization is required. My recollection was that SQL Server 2000 employed a method that ensured even distribution by having each thread process one page at a time(?) This was fine in the Pentium II/III 500MHz generation, but was far too small by the NetBurst and Core2 architecture processors.

    SQL Server 2005 employed a different methods with the strategy of allowing larger granularity and reduced need for synchronization between threads(?) but could also result in having highly uneven distribution of work between threads. Some of this was reported to have been fixed in SQL Server 2008 sp1, see Using Star Join and Few-Outer-Row Optimizations to Improve Data Warehousing Queries but apparently this is still an issue in SQL Server execution plans?

    In the COUNT and no TOP queries, the Constant Scan operation acts as the outer source in a loop join to Document. This is the artificial rowset from the IN clause on AccessControlListID_D. It is known to SQL Server from statistics that this column has highly skewed distribution resulting in the row-thread split shown below.

    The distribution on the other joins are also skewed but not as strongly?

    The query with the TOP clause and an index hint just happens to prevent the SQL Server query optimizer from using a Constant Scan as a loop join outer source. The index hint does nothing to improve the CPU efficiency of this query. In fact it is more than 10% less efficient in terms of CPU. The positive effect is that work is evenly distributed among threads as shown below resulting in nearly linear scaling with parallelism (7.6 to 1).

    Technically the problem here is not all few outer rows, but just uneven few outer rows? Microsoft seems to be aware of the problem and has fixed some aspects? but apparently now all. Perhaps the SQL Server engine could implement a flag or hint to avoid execution plans that are sensitive uneven distributions? For now, the only work-around is to look for this situation and take appropriate action.

    UNION in place of AND/OR Combination

    The execution plan that has outstanding results in terms of CPU efficiency is the UNION replacing the OR clause. It was previously documented that the SQL Server query optimizer has difficulty in generating a good execution plan for queries with a combination of AND/OR conditions, but has no problems when an alternative structure is employed.

    There are other forms for which the SQL Server query optimizer does not produce good execution plans. Once these can be catalogued with appropriate alternative SQL expression strategies, there are probably not any searches that cannot be handled.

    COUNT + TOP Architecture

    This actually encompasses several sub-topics. Presumably the purpose of this architecture is to know the exact number of documents that match a search, but also so as to not over-whelm the application server with data. But we should consider that 1) the query with a rowset only has an integer column, 2) even a large case should have no more than tens of millions of documents (a small number in the modern world) and 3) the application server today has many gigabytes of memory. So this is not absolutely necessary?

    There is also a client/application-side work around for this. Simply issue the query to return the identity column for all rows in the search, load the first X into an array, then continue to read but not store the remaining rows.

    Consider the COUNT + TOP architecture. There are 4 possibilities. One is that both queries are inexpensive in which case this is not important. Second is that the COUNT query is expensive but not the TOP. This could happen when many rows match the search, and the TOP clause allows the query to exit quickly.

    Third is that the COUNT is not expensive but the TOP is. This happens when the query optimizer estimates many rows but in fact there are few rows. In this case, an appropriate high-volume parallel plan is employed for the COUNT query, but a non-parallel plan is used for the TOP query relying on the expectation of exiting quickly. The exit condition is never met, and the non-parallel plan must process the full set of rows with only a single thread. Consider also that this non-parallel plan was formulated based on low start-up overhead (loop joins) rather than volume efficiency (hash joins).

    The fourth possibilities is that both the COUNT and TOP queries are inherently expensive. In this case, we now have to execute two expensive queries all so that the developer can avoid a few lines of code on the application side? (Many SQL/database disasters have been traced to lazy/incompetent coders.)

    Summary

    As I said in the first Relativity article, all of the database problems appear to be solvable, but most require action in the application code where the SQL is generated, including the architectural strategy of COUNT + TOP. Some problems from Relativity 7.x appear to have been resolved, such as the data type mismatch between a varchar column and nvarchar search parameter. This by itself was a nuisance, but when combined with the TOP clause had significant negative consequence.

    The full set of details along with a test database and queries to both reproduce and fix the problems in Relativity 7.x were sent to kCura. Most of the advice seems to have been disregarded.

    It would seem that kCura was aware that there were problems. In version 7.x, there was a CodeArtifact table (3 columns CodeTypeID, CodeArtifactID, AssociatedArtifactID) that was frequently involve in search queries that would take forever, as in 30min (or whatever the web page time-out is) to 30 months (estimated) to complete. It is possible that the long running read queries and write activity also resulted in blocking and deadlocks despite prolific use of NOLOCK (without the WITH keyword?).

    For version 8.x kCura went to great effort to split this table into multiple tables with names of the form ZCodeArtifact_xxx, one for each of the group values (xxx)? The new tables have columns CodeArtifactID and AssociatedArtifactID, so perhaps there is one table for each CodeTypeID in the version 7.x table? (This topic is covered in ZCodeArtifact & Statistics, as there were a series of performance problems in queries to these tables related to statistics.)

    The problem was that blocking and deadlocks were the symptoms of Relativity problems, not the cause. The causes were the topics discussed here: 1) injudicious use of the TOP clause in situations in which the query optimizer makes a serious error in the estimated number of rows when the actual row count is already known from the COUNT query, 2) generating a complicated SQL expression with combination AND/OR clauses instead of JOIN and UNION, and 3) ineffective parallel execution plans due of skewed distribution. One more point, kCura also went to the effort of changing the SQL a from single expression form to one using CTEs. This may contribute to the clarity of the expression, but does not impact the execution plan problems.

    Another problem seen in 8.1 is a search involving both conventional search arguments and Full-Text elements. The execution plan had the Full-Text Search (FTS) operation as a loop join inner source (see FTS), probably because the estimate number of rows from the outer source showed 1 (which could mean 0). This might have been because the ZCodeArtifact_xxx table was newly generated and statistics were not updated? The query produced a good execution plan after statistics were updated and had quick actual execution time as well.

    Appendix

     
    (232 row(s) affected) SQL Server Execution Times: CPU time = 152148 ms, elapsed time = 152298 ms. TOP 1000
    (232 row(s) affected) SQL Server Execution Times: CPU time = 143973 ms, elapsed time = 71857 ms. no TOP
    (232 row(s) affected) SQL Server Execution Times: CPU time = 167671 ms, elapsed time = 22016 ms. Index hint
    SQL Server Execution Times: CPU time = 1665 ms, elapsed time = 324 ms. UNION
    SQL Server Execution Times: CPU time = 138279 ms, elapsed time = 69109 ms. COUNT

    The full list of Document.AccessControlListID_D and o1000022_f2154932.ArtifactID values are below
    /* [AccessControlListID_D]
    1,1000062,1000063,1000064,1000065,1000066,1000067,1000099,1000100,1000101,1000102,1000103,1000104,1000105, 1000106,1000107,1000108,1000109,1000110,1000111,1000112,1000113,1000114,1000115,1000116,1000117,1000118, 1000119,1000120,1000121,1000122,1000123,1000124,1000125,1000126,1000127,1000128,1000129,1000130,1000131, 1000132,1000133,1000134,1000135,1000136,1000137,1000138,1000139,1000140,1000141,1000142,1000143,1000144, 1000145,1000146,1000147,1000148,1000149,1000150,1000151,1000152,1000153,1000154,1000155,1000156,1000157, 1000158,1000159,1000160,1000161,1000162,1000163,1000164,1000165,1000166,1000167,1000168,1000169,1000170, 1000171,1000172,1000173,1000174,1000175,1000176,1000177,1000178,1000179,1000180,1000181,1000182,1000183, 1000184,1000185,1000186,1000187,1000193,1000206,1000240,1000244,1000245,1000246,1000259,1000288,1000315, 1000483,1000535,1000536,1000550,1000599,1000617,1000763,1000770

    [o1000022_f2154932].[ArtifactID]
    22552945,22552935,22552925,22552915,22552905,14768277,14768265,5375743,5375742,5375741,5375740,22552947, 22552937,22552927,22552917,22552907,22552894,22552884,22552933,22552923,22552913,22552903,22552890, 22552880,14768269,5375739,5375738,22552951,22552941,22552931,22552921,22552911,22552901,22552896, 22552886,14768275,22552892,22552882,22552939,22552929,22552919,22552909,14768273,14768267,22552949, 28315630,5549264,22552953,22552898,22552888,22552878,22552954,22552944,22552934,22552924,22552914, 22552904,22552950,22552940,22552930,22552920,22552910,22552900,14768271,22552948,22552938,22552928, 22552918,22552908,22552952,22552942,22552946,22552936,22552926,22552916,22552906,22552895,22552885, 22552932,22552922,22552912,22552902,22552891,22552881,14768268,22552943,22552899,22552889,22552879, 14768270,5549202,22552897,22552887,14768274,14768272,22552893,22552883,14768266,14768276,14768264

    Note: a Test database was created with just these tables and limited columns for in-depth investigation.

More Posts Next page »

This Blog

Syndication

Powered by Community Server (Commercial Edition), by Telligent Systems
  Privacy Statement