Hi,
In an attempt to reduce the high IO usage of the mon, I tried to
enable the LevelDB snappy compression.
The good news is that it did reduce the IO substantially (~ 3.5x less
IO) and also the disk space (by about the same ratio).
Unfortunately, it seems it also takes a LOT of CPU. ( ~ 50/60% of one
core on some Xeon server). And I'm not exactly sure why.
I write a quick snappy benchmark and it compresses on that server at
nearly 500 MB/s and decompresses even faster (all single threaded
test).
The test data I used was a binary PGMap I took from the cluster (so I
get data with same patterns as in reality).
Cheers,
Sylvain
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
