THE SQL Server Blog Spot on the Web
Just a quick note that my CTP3 paper has finally been published, and now I can start working on the final version!
Here’s the link:
If you would like to receive an email when updates are made to this post, please register here
Subscribe to this post's comments using
Are you in a position to disclose the exact tests supporting the claim that the in-memory OLTP engine is scalable up to four CPU sockets. Are you also in a position to disclose how either you or Microsoft worked around the pressure on key spinlocks as your tests neared the loading of all CPU cores across all CPU sockets.
By "Efficient scaling" do you mean that the transaction throughput curve was a straight line as the workload was scaled up ?.
Nice, and I want to read Real-Time Operational Analytics version, that in-memrory OLTP + non-clustered Columnstore index seems to resolve some "Unsuitable Application scenario".
Good stuff, thanks again for your work Kalen!
What if I just answered 'no'? :-)
I don't run the performance tests, Microsoft does, in their labs with many different customers. I am not involved with the testing so I just take the results they give me. Anything I can test for myself, as far as functionality, I will always verify before publishing, but for things that need a lab, or an application and data that will truly benefit from in-memory OLTP, I have to leave that to Microsoft.
could should be at least equal to the number of distinct values for the index key columns." You've used two verbs "could should".
Thanks Chris! The 'could' should have been 'count' and it has already been corrected as I update the paper for RTM. Although it is now on page 84, not 73. :-)
The whole sentence is: "The bucket count should be at least equal to the number of distinct values for the index key columns."
Any estimates on the next edition of the MSSQL Internals book? It's 'bout time :)
Hi Dr. Nick
There will not be any more of the big SQL Server Internals books. The last one took over a year to write, and the publisher didn't get it published for over a year after that. And with a new version coming out every couple of years, I can't update the book fast enough to keep up with that. So I will be doing more smaller internals books, like my Concurrency and In-Memory OLTP books for Redgate.
Sad to hear but understandable. I love reading your "deep dive" books... very useful. Looking forward to new printed material from you.
Thanks for replying!