Too many mutex locks

Too many mutex locks

Post by Steve K » Thu, 12 Dec 1996 04:00:00

After running the SE2.5 performance toolkit on one of our heavily loaded
SPARCserver 1000's, I have learned that it's got a high number of mutex
locks.  Apparently the norm is around 200, but it's peaking at around

- Is there anything I can use to see which PID is responsible?

- Are there any changes that can be made to the system to improve the
  mutex locks?  Perhaps some changes to /etc/system?

- Would adding an extra processor board just make matters worse, increasing
  the mutex locks even more?

Machine info: 320MB RAM, Baydel disk array, 2 processors, running Sybase v10.

Thanks for any help,
Steve Kay