I bet you'd see better memstore results with my vector based object
implementation instead of bufferlists.
Where can I find it?
Nick Fisk noticed the same
thing you did. One interesting observation he made was that disabling
CPU C/P states helped bluestore immensely in the iodepth=1 case.
On 3/12/19 8:40 AM, vita...@yourcmc.ru wrote:
One way or another we can only have a single thread sending writes to
rocksdb. A lot of the prior optimization work on the write side was
to get as much processing out of the kv_sync_thread as possible.
That's still a worthwhile goal as it's
One way or another we can only have a single thread sending writes to
rocksdb. A lot of the prior optimization work on the write side was
to get as much processing out of the kv_sync_thread as possible.
That's still a worthwhile goal as it's typically what bottlenecks with
high amounts of
On 3/12/19 7:31 AM, vita...@yourcmc.ru wrote:
Decreasing the min_alloc size isn't always a win, but ican be in some
cases. Originally bluestore_min_alloc_size_ssd was set to 4096 but we
increased it to 16384 because at the time our metadata path was slow
and increasing it resulted in a pretty
Decreasing the min_alloc size isn't always a win, but ican be in some
cases. Originally bluestore_min_alloc_size_ssd was set to 4096 but we
increased it to 16384 because at the time our metadata path was slow
and increasing it resulted in a pretty significant performance win
(along with
Am 06.03.19 um 14:08 schrieb Mark Nelson:
>
> On 3/6/19 5:12 AM, Stefan Priebe - Profihost AG wrote:
>> Hi Mark,
>> Am 05.03.19 um 23:12 schrieb Mark Nelson:
>>> Hi Stefan,
>>>
>>>
>>> Could you try running your random write workload against bluestore and
>>> then take a wallclock profile of an
On 3/6/19 5:12 AM, Stefan Priebe - Profihost AG wrote:
Hi Mark,
Am 05.03.19 um 23:12 schrieb Mark Nelson:
Hi Stefan,
Could you try running your random write workload against bluestore and
then take a wallclock profile of an OSD using gdbpmp? It's available here:
On 3/5/19 4:23 PM, Vitaliy Filippov wrote:
Testing -rw=write without -sync=1 or -fsync=1 (or -fsync=32 for batch
IO, or just fio -ioengine=rbd from outside a VM) is rather pointless -
you're benchmarking the RBD cache, not Ceph itself. RBD cache is
coalescing your writes into big sequential
Hi Mark,
Am 05.03.19 um 23:12 schrieb Mark Nelson:
> Hi Stefan,
>
>
> Could you try running your random write workload against bluestore and
> then take a wallclock profile of an OSD using gdbpmp? It's available here:
>
>
> https://github.com/markhpc/gdbpmp
sure but it does not work:
#
Testing -rw=write without -sync=1 or -fsync=1 (or -fsync=32 for batch IO,
or just fio -ioengine=rbd from outside a VM) is rather pointless - you're
benchmarking the RBD cache, not Ceph itself. RBD cache is coalescing your
writes into big sequential writes. Of course bluestore is faster in
Hi Stefan,
Could you try running your random write workload against bluestore and
then take a wallclock profile of an OSD using gdbpmp? It's available here:
https://github.com/markhpc/gdbpmp
Thanks,
Mark
On 3/5/19 2:29 AM, Stefan Priebe - Profihost AG wrote:
Hello list,
while the
Am 05.03.19 um 10:05 schrieb Paul Emmerich:
> This workload is probably bottlenecked by rocksdb (since the small
> writes are buffered there), so that's probably what needs tuning here.
while reading:
This workload is probably bottlenecked by rocksdb (since the small
writes are buffered there), so that's probably what needs tuning here.
Paul
--
Paul Emmerich
Looking for help with your Ceph cluster? Contact us at https://croit.io
croit GmbH
Freseniusstr. 31h
81247 München
www.croit.io
Tel:
13 matches
Mail list logo