From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Nelson Subject: Re: First attempt at rocksdb monitor store stress testing Date: Fri, 01 Aug 2014 12:41:55 -0500 Message-ID: <53DBD163.5000004@inktank.com> References: <53D041D3.3080203@inktank.com> <75674D092A819E4189E91166C74CB90D013F2ACA@shsmsx102.ccr.corp.intel.com> <53D0EA63.10107@inktank.com> <53D19A92.3010000@inktank.com> <75674D092A819E4189E91166C74CB90D013F30D4@shsmsx102.ccr.corp.intel.com> <53D28145.2020504@inktank.com> <75674D092A819E4189E91166C74CB90D013F3CD6@shsmsx102.ccr.corp.intel.com> <53D68096.1060009@inktank.com> <75674D092A819E4189E91166C74CB90D01402A44@shsmsx102.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-ie0-f180.google.com ([209.85.223.180]:56177 "EHLO mail-ie0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750774AbaHARmA (ORCPT ); Fri, 1 Aug 2014 13:42:00 -0400 Received: by mail-ie0-f180.google.com with SMTP id at20so6101550iec.25 for ; Fri, 01 Aug 2014 10:41:59 -0700 (PDT) In-Reply-To: <75674D092A819E4189E91166C74CB90D01402A44@shsmsx102.ccr.corp.intel.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: "Shu, Xinxin" , "ceph-devel@vger.kernel.org" On 07/30/2014 09:00 PM, Shu, Xinxin wrote: > Does your report base on wip-rocksdb-mark branch? Yes, though I've been tweaking Joao's test tool a bit. I ran more tests with the higher ulimit for rocksdb, and also did 10,000 objects instead of 5,000. There are some interesting effects. Leveldb appears to be faster for reads, but compaction causes horrible predictable stalls. rocksdb with leveled compaction appears to reduce and/or eliminate the length and duration of the compaction stalls. Surprisingly the spinning disk test had no high outliers while the SSD test had a few. Universal Compaction seems to fair poorly with more objects. The maximum latencies are still bound, but the averages are significantly worse. http://nhm.ceph.com/mon-store-stress/Monitor_Store_Stress_Medium_Tests.pdf Next I will need to examine the test parameters and run much longer tests. Switch to GNUPlot as well. Open office can't handle this much data. :) Mark > > Cheers, > xinxin > > -----Original Message----- > From: ceph-devel-owner@vger.kernel.org [mailto:ceph-devel-owner@vger.kernel.org] On Behalf Of Mark Nelson > Sent: Tuesday, July 29, 2014 12:56 AM > To: Shu, Xinxin; ceph-devel@vger.kernel.org > Subject: Re: First attempt at rocksdb monitor store stress testing > > Hi Xinxin, > > Thanks, I'll give it a try. I want to figure out what's going on in Rocksdb when the test stalls with leveled compaction. In the mean time, here are the test results with spinning disks and SSDs: > > http://nhm.ceph.com/mon-store-stress/Monitor_Store_Stress_Short_Tests.pdf > > Mark > > On 07/27/2014 11:45 PM, Shu, Xinxin wrote: >> Hi mark, >> >> I tested this option on my setup , same issue happened , I will dig into it , if you want to get info log , there is a workaround, set this option to none: >> >> Rocksdb_log = "" >> >> Cheers, >> xinxin >> >> -----Original Message----- >> From: Mark Nelson [mailto:mark.nelson@inktank.com] >> Sent: Saturday, July 26, 2014 12:10 AM >> To: Shu, Xinxin; ceph-devel@vger.kernel.org >> Subject: Re: First attempt at rocksdb monitor store stress testing >> >> Hi Xinxin, >> >> I'm trying to enable the rocksdb log file as described in config_opts using: >> >> rocksdb_log = >> >> The file gets created but is empty. Any ideas? >> >> Mark >> >> On 07/24/2014 08:28 PM, Shu, Xinxin wrote: >>> Hi mark, >>> >>> I am looking forward to your results on SSDs . >>> rocksdb generates a crc of data to be written , this cannot be switch off (but can be subsititued with xxhash), there are two options , Option. verify_checksums_in_compaction and ReadOptions. verify_checksums, If we disable these two options , i think cpu usage will goes down . If we use universal compaction , this is not friendly with read operation. >>> >>> Btw , can you list your rocksdb configuration? >>> >>> Cheers, >>> xinxin >>> >>> -----Original Message----- >>> From: Mark Nelson [mailto:mark.nelson@inktank.com] >>> Sent: Friday, July 25, 2014 7:45 AM >>> To: Shu, Xinxin; ceph-devel@vger.kernel.org >>> Subject: Re: First attempt at rocksdb monitor store stress testing >>> >>> Earlier today I modified the rocksdb options so I could enable universal compaction. Over all performance is lower but I don't see the hang/stall in the middle of the test either. Instead the disk is basically pegged with 100% writes. I suspect average latency is higher than leveldb, but the highest latency is about 5-6s while we were seeing 30s spikes for leveldb with levelled (heh) compaction. >>> >>> I haven't done much tuning either way yet. It may be that if we keep level 0 and level 1 roughly the same size we can reduce stalls in the levelled setups. It will also be interesting to see what happens in these tests on SSDs. >>> >>> Mark >>> >>> On 07/24/2014 06:13 AM, Mark Nelson wrote: >>>> Hi Xinxin, >>>> >>>> Thanks! I wonder as well if it might be interesting to expose the >>>> options related to universal compaction? It looks like rocksdb >>>> provides a lot of interesting knobs you can adjust! >>>> >>>> Mark >>>> >>>> On 07/24/2014 12:08 AM, Shu, Xinxin wrote: >>>>> Hi mark, >>>>> >>>>> I think this maybe related to 'verify_checksums' config option >>>>> ,when ReadOptions is initialized, default this option is true , >>>>> all data read from underlying storage will be verified against >>>>> corresponding checksums, however, this option cannot be >>>>> configured in wip-rocksdb branch. I will modify code to make this option configurable . >>>>> >>>>> Cheers, >>>>> xinxin >>>>> >>>>> -----Original Message----- >>>>> From: ceph-devel-owner@vger.kernel.org >>>>> [mailto:ceph-devel-owner@vger.kernel.org] On Behalf Of Mark Nelson >>>>> Sent: Thursday, July 24, 2014 7:14 AM >>>>> To: ceph-devel@vger.kernel.org >>>>> Subject: First attempt at rocksdb monitor store stress testing >>>>> >>>>> Hi Guys, >>>>> >>>>> So I've been interested lately in leveldb 99th percentile latency >>>>> (and the amount of write amplification we are seeing) with leveldb. >>>>> Joao mentioned he has written a tool called mon-store-stress in >>>>> wip-leveldb-misc to try to provide a means to roughly guess at >>>>> what's happening on the mons under heavy load. I cherry-picked it >>>>> over to wip-rocksdb and after a couple of hacks was able to get >>>>> everything built and running with some basic tests. There was >>>>> little tuning done and I don't know how realistic this workload is, >>>>> so don't assume this means anything yet, but some initial results are here: >>>>> >>>>> http://nhm.ceph.com/mon-store-stress/First%20Attempt.pdf >>>>> >>>>> Command that was used to run the tests: >>>>> >>>>> ./ceph-test-mon-store-stress --mon-keyvaluedb >>>>> --write-min-size 50K --write-max-size 2M --percent-write 70 >>>>> --percent-read 30 --keep-state --test-seed 1406137270 --stop-at >>>>> 5000 foo >>>>> >>>>> The most interesting bit right now is that rocksdb seems to be >>>>> hanging in the middle of the test (left it running for several >>>>> hours). CPU usage on one core was quite high during the hang. >>>>> Profiling using perf with dwarf symbols I see: >>>>> >>>>> - 49.14% ceph-test-mon-s ceph-test-mon-store-stress [.] >>>>> unsigned int >>>>> rocksdb::crc32c::ExtendImpl<&rocksdb::crc32c::Fast_CRC32>(unsigned >>>>> int, char const*, unsigned long) >>>>> - unsigned int >>>>> rocksdb::crc32c::ExtendImpl<&rocksdb::crc32c::Fast_CRC32>(unsigned >>>>> int, char const*, unsigned long) >>>>> 51.70% >>>>> rocksdb::ReadBlockContents(rocksdb::RandomAccessFile*, >>>>> rocksdb::Footer const&, rocksdb::ReadOptions const&, >>>>> rocksdb::BlockHandle const&, rocksdb::BlockContents*, >>>>> rocksdb::Env*, >>>>> bool) >>>>> 48.30% >>>>> rocksdb::BlockBasedTableBuilder::WriteRawBlock(rocksdb::Slice >>>>> const&, rocksdb::CompressionType, rocksdb::BlockHandle*) >>>>> >>>>> Not sure what's going on yet, may need to try to enable >>>>> logging/debugging in rocksdb. Thoughts/Suggestions welcome. :) >>>>> >>>>> Mark >>>>> -- >>>>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" >>>>> in the body of a message to majordomo@vger.kernel.org More >>>>> majordomo info at http://vger.kernel.org/majordomo-info.html >>>>> >>>> >>> >> > > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html >