http://blogs.nologin.es/slopez/archives/17-Benchmarking-Databases-I.-Volatile-Storage..html PostgreSQL, MySQL and Oracle are compared using reasonable hardware setup and TPC-C like test with BenchmarkSQL tool (JDBC ) . This post will probably disappear soon as the author clearly violates oracle license that does not permit publishing benchmark results w/out permissions
http://www.mysqlperformanceblog.com/2006/08/09/cache-performance-comparison/ gives you an idea how different caching methods perform. Read it before implementing something
http://archives.postgresql.org/message-id/[email protected] Dell 2950
http://merlinmoncure.blogspot.com/2007/08/following-are-results-of-our-testing-of.html bonnie++ benchmark comparing dell perc 5 (lsi) hw raid controllers connected to Dell md1000 jbod box and linux md raid driver. (test was performed by a notable PostgreSQL developer )
http://tservice.net.ru/~s0mbre/old/?section=projects&item=netchannel Network channel is a protocol agnostic communication channel between hardware and userspace. Basically it allows very fast delivery from network hardware to userspace All protocol processing happens in process’ context.
http://revsys.com/writings/postgresql-performance.html yet another postgresql tuning guide, contains useful explanation of most omportant config parameters
http://momjian.us/main/writings/pgsql/hw_performance/index.html notes on PostgreSQL tuning from Bruce Momjian (PostgreSQL core team )
http://blog.emmettshear.com/post/2008/03/03/Dont-use-Pound-for-load-balancing another evidence that if u need a loadbalancer – ngnix is the answer these days [and maybe perlball]
http://www.infoq.com/articles/ebay-scalability-best-practices the subj says it. good article
http://people.planetpostgresql.org/jkshah/index.php?/archives/3-Problems-with-PostgreSQL-on-Multi-core-Systems-with-Multi-Terabyte-Data.html PostgreSQL: benchmarks and further analisys for OLTP workloads 8-32 cores with ~TB db size from Sun’s potgresql team