Mark has found Double sysbench throughput with TCMalloc with is quite amazing.
In Drizzle, if TCMalloc is present, we’ve linked it in (and on Solaris, mtmalloc) for years now. well…. Drizzle years, which are rather like dog years.
Mark has found Double sysbench throughput with TCMalloc with is quite amazing.
In Drizzle, if TCMalloc is present, we’ve linked it in (and on Solaris, mtmalloc) for years now. well…. Drizzle years, which are rather like dog years.
Fix the HEAP engine and the results won’t be as amazing. When will someone publish similar results for Solaris to compare glibc malloc with mtmalloc? Has drizzle eliminated LOCK_open and the access control mutex? Or just LOCK_open?
hrrm… drizzle uses a different heap engine, so that could help a lot too.
I have linux on a T1000… but no solaris on it… if somebody has a T1k with solaris, we could compare.
We still have LOCK_open in a bunch of places – but it should be able to go away with my further work.
acl mutex should be gone (no acl)