SQL Server Days 2016 - lessons learned
Some weeks ago, I had the pleasure to speak at the Belgian SQL Server conference SQL Server Days
I’ve been a busy bee at work and in my private life, and haven’t had the time to write up a blog post on lessons learned. But better late than never, so here goes…
I work for Microsoft, and hear about lots of awesome, interesting (and secret) things, which I cannot share here in my private blog. But the following was disclosed in public sessions, and I assume that all of the information is free of NDA.
SQL Server on Linux will be in public preview in December 2016. I could not find any public statements on this
Gartner’s 2016 Magic Quadrant for Operational Database Management Systems has been published, and have placed Microsoft as the industry Leader ahead of Oracle and all other vendors with the highest scores on both vision and execution https://www.gartner.com/doc/reprints?id=1-3JD7HF0&ct=161005&st=sb
The Azure team are working on a feature called “Workload Insight.” Might be reporting on top of Query Store. I could not find any public statements on this
A new Fast Track specification for SQL Server 2016 has been made public (mostly relevant for data warehousing guys) https://www.microsoft.com/en-us/cloud-platform/data-warehouse-fast-track
The team around Polybase are working on integrating new database types into the engine. I could not find any public statements on this
SQL Server 2016 has been improved to be able to utilize persistent memory mapped as storage in Windows Server 2016. Especially for the transaction log, this can really speed things upd (for systems with waits on write to the logs). See this video for more details: https://channel9.msdn.com/Shows/Data-Exposed/SQL-Server-2016-and-Windows-Server-2016-SCM--FAST
The new Gartner buzzword for running analytics directly on top of OLTP systems is now called Hybrid Transactional/Analytical Processing (HTAP), see https://en.wikipedia.org/wiki/Hybrid_Transactional/Analytical_Processing_(HTAP)
Schema changes for attribute datatype int->bigint can now be done as an online operation. I could not find any public statements on this
If a table is row/page compressed, schema changes for attribute datatype int->bigint can now be done as metadata operation. I could not find any public statements on this
It is possible to truncate a table partition in one operation. Truncate partition will require an exclusive schema (SCH-M) lock on the table. See MSDN for more info: https://msdn.microsoft.com/en-GB/library/ms177570.aspx
Personally, a gave the talk WTF with SQL, with topics from set theory, relational algebra, NULL and the SQL standard, as well as implementation of the standard in SQL Server.
I had to leave the conference early to catch a flight to Prague, to attend the NAV Directions EMEA 2016 conference, so I didn’t have a lot of time talking to old and getting new friends from the #SQLFamily, talking to sponsors, and networking with speakers. Hopefully next time…