…reads, vCache with time-based flushing consistently hit about 27000 per second, whereas without time-based flushing it averaged about 12500. FlashCache…–rand-type=pareto \ –forced-shutdown=1 –max-time=7200 –max-requests=0 –percentile=95 \ –mysql-user=root –mysql-socket=/tmp/…
Post: MySQL and Percona Server in LinkBench benchmark
May 8, 2013 By 18 Comments
… result. In the best possible combination I got ~15k operations per second for MySQL 5.6.11. Conclusion: In CPU-bounds case… line to run test for 10x dataset: ./bin/linkbench -D requesters=64 -D dbid=linkdb -D host=127.0.0.1….properties -csvstats final-stats.csv -csvstream streaming-stats.csv -D requests=5000000 -D maxtime=1800 -r
Post: More on MySQL transaction descriptors optimization
April 26, 2013 By Leave a Comment
… single socket x 2 threads per socket), while the second one use 16 cores (8 cores per socket per node). So this test also… –rand-type=uniform –forced-shutdown=1 –max-time=120 –max-requests=0 –percentile=99 –mysql-user=root –mysql-db=sbtest8t1M –mysql… –rand-type=uniform –forced-shutdown=1 –max-time=120 –max-requests=0 –percentile=99 –mysql-user=root –mysql-db=sbtest8t1M –mysql…
Post: Configuring MySQL For High Number of Connections per Second
January 6, 2012 By 23 Comments
… that there were roughly 2,000 new connections to MySQL per second during peak times. This is a high number by any… and 2000 connections created per second it will take just 0.025 seconds to fill the queue completely if requests are not served, which…). Another important setting if you’re working with many connections per second is thread_cache which impacts cost of connections dramatically. You…
Post: Evaluating IO subsystem performance for MySQL Needs
March 5, 2008 By 23 Comments
… interested about writes which show us how many cached writes per second we can do – this is important for log writes, which… many outstanding requests. Running with more than one threads you may also take a look at another table SysBench displays: per-request statistics… –file-test-mode=rndwr run Now we get 827.94 Requests/sec requests per second, which is not bad considering this is RAID5 volume…
Post: Virtualization and IO Modes = Extra Complexity
March 21, 2011 By 12 Comments
… to do no more than a bit over 100 writes per second as even in case there is no disk seek we… as it times only writes not fsync() calls…. 22 fsync requests per second is pretty bad though I assume It can be realistic… events: 33900 total time taken by event execution: 37.5364 per-request statistics: min: 0.10ms avg: 1.11ms max: 259.69ms…
Post: The two even more fundamental performance metrics
May 5, 2011 By 6 Comments
… be a steady but variable “white noise” of incoming requests, and most requests will take about the same time to process and… the server handles 20k requests per second. A few of the requests will arrive in one second and complete in the next second, so at the… of requests that span the transition between seconds will be small. So if we count the arrivals and completions in each second, and…
Comment: Lighttpd as reverse proxy
June 19, 2008 By 28 Comments
… seconds Complete requests: 100000 Failed requests: 0 Write errors: 0 Total transferred: 254900000 bytes HTML transferred: 224100000 bytes Requests per second: 9134.93 [#/sec] (mean) Time per request: 21.894 [ms] (mean) Time per request: 0.109 [ms] (mean, across all concurrent requests) Transfer…
Post: When would you use SAN with MySQL ?
March 9, 2009 By 18 Comments
… requests per second. True the SAN can have much more hard drives than directly attached storage but usually also at much higher prices per… need to be serialized. Second, MySQL is often not able to submit a lot of outstanding IO requests which is needed to… usually it is not able to drive number of outstanding request significantly. For write intensive workload you will have problem with…
Post: Innodb Fuzzy checkpointing woes
May 10, 2006 By Leave a Comment
… data – 100% “utiization” corresponds both to 1 or 20 outstanding requests which are very different. One however probably can combine… 1000 requests per second and the best would be to keep 16 outstanding requests at the time.  The number of outstanding requests is also very important - if you do not submit enough concurrent requests IO subsystems will not…
