1

I've an SSIS package with the following design:

  1. There're 20 data flow tasks in a package that read data from the same table and load to 20 different tables.
  2. MaxConcurrentExecutable is set to 6, so it executes 6 tasks in parallel.
  3. There is a single connection manager for all source components with RetainSameConnection is set to true.
  4. There are 20 additional connection managers used by each respective destination.
  5. Source and destination are on the same SQL Server database. SSIS runs on a separate server. Both database and SSIS servers are 4cpu and 8gb ram.

This is running very slowly. When I see sp_who2, I notice the select queries (spids) that read from the source are suspended sometimes blocked by themselves, and spids that do bulk insert to destination are awaiting command.

Can anyone tell me why this is happening. Where do I need to look to optimize this (database performance, hardware, etc.,)?

I appreciate your help. Thanks!

4

0 回答 0