|
|
|
It was reported that FairCom Server was hitting a throughput limit due to the overhead imposed by internal lock counters (e.g., ctstat.exe -filelocks, SnapShot(), etc.). These counters used a strong mutex when updating lock statistics.
Atomic operations are much more efficient for these global update counters and are now used instead. In the reported case, performance gains after this change were close to doubling throughput (approximately 33K TPS to nearly 65K TPS).
Note: When this optimization is enabled, which is on by default, the FairCom Server LOCK_MONITOR keyword is no longer available. The lock statistics available through ctstat and SnapShot() are still available and are more complete than the LOCK_MONITOR keyword.
Note: This change is a server‑side only change and it is not mandatory to relink your client applications to benefit. However, it is FairCom's best practice recommendation to always relink all client‑side applications such that the client version always matches the FairCom Server process whenever possible.
