Re: [infinispan-dev] Per-entry lock container
On 22.10.2012 18:16, Manik Surtani wrote:
Wow. Looks like ref counting is a winner. Simple local mode test,
comparing 5.2.0.Beta2 (in red), my fix for ISPN-2381 *without* ref
counting (in blue) and with ref counting (in green).
Weird that read performance is also affected, as these patches are for
the lock manager which is not used when reading. Could be overall
resource contention that slows down reader threads.
[...]
Thats just in time. I noticed two weeks ago during internal benchmarks,
that the current exclusive locking stuff for write operations in local
caches is expensive.
Looking forward to have this: pull, pull, pull! :-D
Attachments:
- attachment.html (text/html — 1.2 KB)