One, Two, Three… Boom!


...I am back and getting ready to start blogging.

It has been almost ten years since my last blog post. Time flies, indeed!  As you can imagine I have plenty of things to share many of which are related to SQL Server, runtimes and  operating systems in general.

As we expanding our horizons, I am also planning  on taking many of you for a "design ride".  Yes, you got it right. My plan is to leverage your experience to help us with making design choices and bring you the best product possible.

On the side note, I know many of you are wondering, what I have been doing for the last ten years.  In brief, I moved to Midori project in the early 2007. I had spent seven years working there. There I worked on kernel and parts of the runtime. As many of Midorians I had had blast. I had really enjoyed working there. It was very special. Then after Midori, I had taken a quick few months d-tour to work on  Windows kernel team, specifically I had worked on Hypervisor for short period of time. In early 2015, I moved back to SQL Server to work on project Helsinki with the goal of bringing SQL Server to Linux.  And now you can enjoy some results of our work.  Today, you can play with SQL Server on Linux.  While in SQL, my role has expanded and I work on way more things that I could have ever imaged.  And yes, I will share more on this in the future...

So, now lets do the real stuff: Do you have questions? If you do then ask!

The game is back on! Lets have fun.


Comments (4)

  1. Niels Berglund says:

    Welcome back Slava! It’s been a while!!!

  2. Steve Ash says:

    I had this blog in my feedly (and previously google reader) since the last time you blogged! So nice to hear from you again! Since you asked for questions and have an expanded role — I have two unrelated, minor questions if you know the answers:

    1- is it still true that the optimizer assumes all I/O require a physical I/O? I.e. there is no model of the probability of a buffer cache hit for a particular data structure to shade the cost when it believes that most of the logical I/Os wont incur real physical I/Os?

    2- does the variable length for fields impact memory sizing of working memory during query execution? that is– if i have a table with a varchar(100) but all of the values in that column are < 10 chars. If I up that to a varchar(4000) — is there _any_ impact at query execution time like maybe reserving more memory to service the query?

  3. Welcome back Slava..Looking forward to your blogs!

  4. AJ says:

    Hi Slavo, been a big fan of your work right from the Windows days. Quick question about the SQL on Linux part. How soon are we planning to catchup on the Advancements in this space. I know this is very early stages, but i would love to see more power to SQL Server as much as we can do on Windows Ecosystems today. Not looking for a direct answer, but only here to seek the advise from the Guru.

Skip to main content