Watch Out For That Denali!

No, not that one. This one!

In case you missed it, last week at the PASS Summit, Microsoft announced the release of the first community technology preview of the next version of SQL Server, code named Denali. It is available for download from Microsoft here. Unfortunately, as the sole DBA at my company, I don’t have the time or spare machines to try out alphas and betas of new software, but I have read about it and some of what I’ve read gets me excited for this version.

Microsoft has put the Denali Book Online up on the web, but, as would be expected at this stage of the game, documentation is sparse and not all functionality is present in the product. But here are a couple of the changes I think will be the most useful for me.

  • Contained databases. Finally, we are approaching truly portable databases. Right now, when you move a database from one server to another, be it via BACKUP / RESTORE or ATTACH / DETACH, there is a bunch of related stuff that doesn’t get transferred – jobs, roles, linked servers, and, of course, logins – which leads to the orphaned user issue that has plagues DBAs for years. A contained database is one that has no dependencies on the instance of SQL Server it is located on. You should just be able to take that database to any other SQL Server and everything will move with it. Aaron Bertrand states in his blog post about this that Denali will only support partially contained databases and support for fully contained databases will be in a future version of SQL Server. Disappointing, but at least it’s a step in the right direction!
  • HADR. Not the super collider. The super availability-er. HADR stands for High Availability and Disaster Recovery and is a new feature of Denali. The Books Online entry is here. This sounds like database mirroring taken to the next level – and with a whole new set of lingo. Instead of the principal and mirror servers that database mirroring uses, HADR uses availability groups, primary replicas, and secondary replicas. I can best explain this using the terminology of mirroring: An availability group is a set of databases that are (using the SQL 2008 terminology), mirrored and can failover as a group. (Nice!) Primary replicas are the “principal” databases in the availability group. Secondary replicas are the mirrored databases on the mirror servers. Ok, the plural servers is a bit of a wish on my part, but the BOL entry states “This CTP supports only a single, asynchronous secondary replica.” which implies multiple secondaries will be supported eventually (although maybe not in the final Denali release). And finally, the other cool thing about HADR is that the secondary replicas will be available for read-only access. Double nice! As you know, with database mirroring, the databases on the mirrored server are in a restoring state and not available for use by end users.
  • Lastly, DTS is history. Yup. No more support for DTS. Start migrating your DTS packages to SSIS now.

Those are just my first impressions based on the limited reading I have done. It’s been a busy week for me, so I haven’t had much time to delve deeper into this. But what I have seen so far excites me!


Leave a Reply

Your email address will not be published. Required fields are marked *

I am a real person and can prove it by doing this math problem: (required) Time limit is exhausted. Please reload CAPTCHA.