Sheeri explains why turning on the query cache without thinking about the consequences is bad but also why simple benchmarks proving 'the query cache is bad' may be skewed

Mark Callaghan tests the query cache. "The query cache has had an interesting history with MySQL. I don't have much experience with it. I might want to use it in the future so I enabled it during a run of sysbench readonly on an 8-core server. The results are fascinating."

At the March 2009 Boston MySQL User Group meeting, Jacob Nikom of MITs Lincoln Laboratory presented Optimizing Concurrent Storage and Retrieval Operations for Real-Time Surveillance Applications. In the middle of the talk, Jacob said he sometimes calls what he did in this application as real-time data warehousing, which was so accurate I decided to give that title to this blog post. The slides can be downloaded in PDF format (1.3 Mb) at http://www.technocation.org/files/doc/Concurrent_database_performance_02.pdf. This talk discussed how to do real-time retrieval operations while doing concurrent high volume insertion, including: * How to keep up with 1.5 Mb/second per server incoming data stream * * server hardware comparison between a multi-core AMD Opteron and a multi core Intel Xeon * MySQL/Postgres comparison * schema design * design of the storage/retrieval benchmark *  []

Recently, I had an opportunity to participate in a stress-testing exercise. By stress-testing, I mean simulating an expected peak load on the database and observing how the database performs. The objective is to make sure the particular hardware can handle the expected load and also to test where the physical limit of the machine lies if we keep increasing the number of concurrent users. I searched for tools available out there. On forums, most people were talking about Hammerora, and although some people recommended high end tools for big bucks, I decided to try this free tool that promised to do exactly what I needed it to do. The open source Hammerora turned out to be excellent tool for the purpose of simulating a typical transactions load in any number of threads. It's built with the Tcl scripting language and it can simulate a real workload with as many client user sessions as your system can handle.So if you're looking for  []

This article explains how to test the performance of a large database with MySQL 5.1, showing the advantages of using partitions. The test database uses data published by the US Bureau of Transportation Statistics. Currently, the data consists of ~ 113 million records (7.5 GB data + 5.2 GB index).