I mentioned earlier that IO scheduler CFQ coming by default in RedHat / CentOS 5.x may be … for MySQL. And yesterday one customer reported that just changing cfq to noop solved their InnoDB IO problems. I ran tpcc… disks in RAID10, controller Perc/6i with BBU) to compare cfq, deadline, noop and anticipatory (last one just to get number…
Post: Fast storage: 8 SSD Intel X-25M 80GB benchmarks
January 18, 2010 By 15 Comments
… point for research, I think there some serialization in Linux software raid or Linux scheduler or on Adaptec hardware level. 2. Cards connected… connected in hardware raid is doing much better – DEADLINE outperforms CFQ here (which is different from random writes) – Software raid0 performed…
Post: Linux IO Schedulers and MySQL
January 30, 2007 By 7 Comments
… about Linux IO Schedulers today which is quite interesting. It goes in details about schedulers and explains in which of workloads which of schedulers… not really significant. This article also points out benchmarking IO schedulers you should look at more numbers than aggregate bandwidth – you… to see the difference. From the article it looks like CFQ should be good choice for databases and it is also…
Comment: FlashCache: tpcc workload
May 26, 2010 By 18 Comments
…, SSD) CFQ just dies vs Deadline in OLTP workloads. See for example http://www.mysqlperformanceblog.com/2009/01/30/linux-schedulers-in-tpcc… just by one command echo deadline > /sys/block/sda/queue/scheduler
Post: Living with backups
November 6, 2008 By 28 Comments
…one-liners) On Linux there is also a utility called ionice. It allows to affect how I/O scheduler will be …cfq” > /sys/block/$device/queue/scheduler ; fi ; done # cat /sys/block/sd?/queue/scheduler noop anticipatory deadline [cfq] noop anticipatory deadline [cfq] noop anticipatory deadline [cfq…
Post: XtraDB benchmarks - 1.5X gain in IO-bound load
December 18, 2008 By 30 Comments
… let me show you why I took NOOP IO scheduler instead of CFQ, there are result for XtraDB with both: 4X difference… to remember as Linux kernels 2.6.18+ (which are used on CentOS / RedHat 5.2) are coming with CFQ scheduler as default. So echo ‘noop’ > /sys/block/sda/queue/scheduler should be one of first things…
Comment: Linux schedulers in tpcc like benchmark
October 27, 2009 By 19 Comments
Stumbled on this thread. I’m benchmarking fs/schedulers on a small HW RAID-5 set of 4x400GB SATA. …, Cached, Adaptive Read-ahead. Kernel 2.6.31.5 (Arch Linux). Very short recap with some numbers (40 simultaneous seq reads…/s (exploded once with corrupt data when changing scheduler) EXT2+noop: 97MB/s XFS+cfq: 30MB/s Some comments on this setup…
Comment: Living with backups
November 18, 2008 By 28 Comments
… trick) though I tested various schedulers/setting do see how it affects the results. Originally running Linux-2.6.22, I had to upgrade to 2.6.27 due to a bug in CFQ: performing read… read operation are fast and MySQL starts lagging slightly. – Using CFQ with default class, side read operation is about twice as…
Post: Performance impact of complex queries
January 17, 2007 By 11 Comments
… milliseconds you would expect disk IO to take. Scheduling Issues OS Disk IO schedules may also “help” in this case. Number of… willing to take. Especially some older disk schedulers could be poor. I remember in Linux 2.4 times I could make single… for directory with couple of files. In recent Linux Kernels deadline or cfq IO schedulers should be better with this. So what can…
Comment: Linux schedulers in tpcc like benchmark
January 30, 2009 By 19 Comments
I have detailed why this happens at http://dammit.lt/2008/02/05/linux-io-schedulers/ – though it all started simply when we compared how I/O behaves on identical machines ( http://dammit.lt/2007/12/13/io-scheduler-deadline-cfq/ )
