Similar to CXPACKET wait type, except this wait will accumulate on controller packets (thread co-ordination) rather than data packets. Occurs when lot of the long running queries are being parallelized. If processor utilisation is very (typically above 80%) this wait should be investigated
DBAs and Developers
- Examine high CPU queries to see if they can be optimised.
- Investigate changing default MAXDOP and parallelism settings (links below)
- Indexing, look for high CPU parallel queries