If you have a request for our blogoverflow blog (http://dba.blogoverflow.com) then please put them here, one per "answer", and someone will work on a blog entry on it, I'm sure.

Also see:

Our blog is launched as of this week! (Feedback here please!)

| | | | | |
  • Does it matter if there are already good blog articles on the subject out there? You guys have helped me understand many things that I've failed to grasp with other blog articles until later – Rachel May 31 '12 at 19:44
  • Not at all, we look to be a repository of helpful information for as much as we may, so we'll take any requests and guest writers, so long as they adhere to the dba.SE mission! – jcolebrand May 31 '12 at 21:05

Something solid about Entity-Attribute-Value (EAV) tables and when or when not to use them

They seem convenient at first glance, however I've had many people tell me not to use them and sending me all over the place to read more about them. It would be nice to have a solid summary about EAV tables somewhere, and when or when not to use them.

| | | | | |
  • Rachel, I apologize in advance if this is one of the posts you've been sent to before, but I wrote this blog post defending EAV to some extent after some similar FUD about how evil the model is. It is definitely a polarizing topic, and I have no intention of blogging about it here - the big challenge is how many variables are involved. Kind of like trying to write generically about when to use CLR or XML when use cases are typically quite specialized and context-dependent. – Aaron Bertrand Jun 9 '12 at 2:09
  • @AaronBertrand Thanks for that :) I wasn't really looking for something that tells me to use or not use the EAV pattern, but instead something outlining pros/cons of it, what to consider when considering using an EAV table, and some example cases of when to use or not use it. Basically, something easy to read that will let me make an informed decision about my own case, with links to other articles containing more detailed information :) – Rachel Jun 9 '12 at 2:15
  • 1
    Also came across this SO question. I'm including these here as much for your benefit as to provide references for the person who ends up writing the blog post. – Aaron Bertrand Jun 9 '12 at 13:05
  • @AaronBertrand Great link, thank you :) – Rachel Jun 11 '12 at 12:58

At my previous employer, we had one database. At my new job, we have almost 300. I'd definitely be interested in reading about any tips or tricks for managing a large number of databases and what you have to do differently from managing just a few.

| | | | | |

Perhaps an article on what CTE's are, and how/when they should be used?

I know there's a decent amount of material out there on CTE's already, however I had trouble understanding exactly what they were for and when to use them over a temp table or table variable until I asked a question on here and got this answer.

A short and simple article on what they are and when to use them would be great :)

| | | | | |
  • I think it's in the works... – JNK Jun 1 '12 at 18:25
  • Common Table Expression would be very good. I would like to see also which databases support CTEs and if/any differences between them will be a good addition. – Atilla Ozgur Jun 2 '12 at 21:08

Perhaps a complete chart or list of all the pros and cons to naming identity columns by either Id or EntityId?

I'd prefer an neutral point of view for this since I don't think there is one right answer to this question, and what you choose probably depends on how you're using the data and writing queries.

For example, as a programmer using an ORM software for database access, I prefer using Id since its easier to use Customer.Id in my code than Customer.CustomerId, and the ORM takes care of most query writing for me, however many dbas prefer CustomerId since then they don't have to use the table alias when writing queries, and there's less chance for an error this way.

A unbiased list of pros/cons of each would be nice to help make an informed decision when deciding on how to name identity fields

Some further reading...

| | | | | |
  • It seems you have a good starting point to writing this blog post yourself :) – ypercubeᵀᴹ Jun 16 '12 at 8:48

One of the new features in SQL Server 2012 is SQL Server Distributed Replay using Profiler. I'd really be interested in reading a post about how this is different from the standard replay and how it might be implemented.

| | | | | |

I'd be very interested to read about how DBAs here do major code releases to production systems. For me, it's an nerve-racking experience involving lots of performance testing, rollback scripts, and waking up in the morning realizing something I've forgotten. But I've never read about another DBA's experience in this. Best-practices would be great.

| | | | | |
  • 1
    For us with three dba's it is a nerve wracking experience involving lots of scripting, rollback preparations and being prepared to wake up in the morning screaming because we forgot something. :-\ (we're working on automation for this, fear not!) – jcolebrand Jun 16 '12 at 7:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .