Thanks all!! Once I take care of some other stuff this morning (and let the coffee start working!) I hope to be able to get back to this issue....

I don't think I see a problem with the two numbers you mention either.

Ninja's_RGR'us (10/17/2010)

---

How about using ROW_NUMBER(PARTITION BY Member_ID, ORDER BY EndDate)

Then self join on ROWNumber = Rownumber + 1 and member_id= member_id

where dateadd(D, 30, tbl1.enddate) <= tbl2.stardate

I'm sure you'll figure out...

amazingly good maintenance stuff here: ola.hallengren.com. You can have it update statistics on indexes that are not defragged at all or are only reorg'd.

Do you have any reference...

Avoid maintenance plans!!!

I HIGHLY recommend the incredible and FREE goodness at ola.hallengren.com for all maintenance activities. Learn to use the goodness there!

I consider Erland's guideance on this topic to be an excellent resource: http://www.sommarskog.se/arrays-in-sql.html

Since you mentioned something changed (merge replication), it is quite possible that blocking or resource contention is at fault (although I still bet some plan caching/parameter sniffing is to blame...

1) disk queue length is a useless measure of IO performance on modern hardware. avg disk sec/read and /write are best.

2) do a fileIO stall analysis and wait stat...

PLEASE do yourself a favor and get a professional to help you configure your IO system correctly. Right now is the ONLY time you will have to do MANY...

1) rapidly out-of-date statistics perhaps?

2) how about doing option recompile on just a few problematic statements in the sproc instead of the entire thing? Usually it is just one...

David Walker-278941 (10/12/2010)

---

I originally didn't like GUIDs because I thought they used too much "space". They really don't. Disk space is cheap, and GUIDs are only 16 bytes...

A/A clusters are a good way to maximize hardware investment. One thing I advise my clients is to have a monitoring agent of some flavor trigger on a failover...

<SmartA$$ ON>I LOVE it when clients use GUIDs in their database schemas - more tuning work for me!! 😀 <SmartA$$ OFF>

We SOOOOOOO need full Windowing Function support in SQL Server. And yes, performance on this type of query will currently be exceptionally poor and approaching non-functional on increasingly large...