May 22, 2012

Post: Benchmarking single-row insert performance on Amazon EC2

… developed by Tokutek. Though the “1 billion row insert challenge” for which this tool was originally built is long over, but…. MySQL Configuration I used Percona Server 5.5.22-55 for the tests. Following is the configuration that I used: ## InnoDB… simulate 5 concurrent connections writing to the table, with each instance of iiBench writing 200 million single row inserts, for a total…

Comment: How much memory Innodb locks really take ?

… row-locks are used by one user to lock records for edit/update, so the transaction is not interrupted until the… a program is worth a 1000 instructions. Please do this for us. I am sure there are many in this situation…

Comment: Testing Fusion-io ioDrive - now with driver 3.1

Rob, Easiest way – just put everything on the Fusion-io card. For more fine tuning it can be something more complex, but it is case-by-case.

Comment: Testing Fusion-io ioDrive - now with driver 3.1

Thanks for posting the benchmarks. They are very informative. One question: when …

Comment: Benchmarking single-row insert performance on Amazon EC2

Any particular reason for choosing 4 1GB log files instead of 2 2GB?

Comment: Percona Server 5.5.23-25.3 released!

@ James Ubuntu Precise (12.04 LTS) will be officially supported with the next Percona Server releases. Percona Server 5.5.24-26.0 is currently scheduled for release in the first week of June.

Comment: Benchmarking single-row insert performance on Amazon EC2

… still a good challenge. The URL for iibench is slightly incorrect. The correct URL for version you used is http://tokutek.com…

Comment: Benchmarking single-row insert performance on Amazon EC2

Interesting results, and well presented. Would you mind sharing the command line you used for your iiBench clients? Also, did you run a test with innodb_flush_log_at_trx_commit = 1?

Comment: Benchmarking single-row insert performance on Amazon EC2

Partitioning probably reduces the stress on the insert buffer. Does XtraDB/Percona Server have an option to use more IO for insert buffer merges when the ibuf gets too big?

Comment: Announcement of Percona XtraDB Cluster 5.5.23

Gilles, you can filter what tables to replicate using regular replication filter i.e. replicate_*, for it does not affect a node propagation. The new node still will copy all databases and instances.