performance with GUID col and large tables

  • We are passing a GUID and an index in a URL. The parameters are passed to a SQL stored procedure for validation. The index is an identity field that is the primary key of table 1. We need to compare the GUID with a GUID stored in table 1 to validate the URL. Then we need to use a third column in table 1 as an index (also identity & primary key for table 2) into table 2 and return the row.

    Is a SELECT with sub-queries preferable to a JOIN? Any suggestions on best performance for larger tables with millions of entries?

    Low overhead on the initial insertion is also important. I think it is better not to make a ROWGUID and not have to index the GUID field either, but am open to suggestions.

    thanks,

    Mark

  • At PASS some of the MS guys mentioned that at around 400/sec, identities become a bottleneck. GUIDs do not as there are routines optimized for their creation. Not sure about indexing.

    Steve Jones

    sjones@sqlservercentral.com

    http://www.sqlservercentral.com/columnists/sjones

    http://www.dkranch.net

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply