Benchmark

http://www.redhat.com/pdf/rhel/bmsql-postgres-sqlsrvr-v1.0-1.pdf “This paper compares the performance of an Online Transaction Processing (OLTP) based

http://www.thebitsource.com/featured-posts/rackspace-cloud-servers-versus-amazon-ec2-performance-analysis/ Testing amazon and rackspace instances: kernel compile time and “how much it costs to compile kernel on $instance”, and IO tests using IOzone – in short: rackspace is cheaper/better, about 2x on average across the board. Note about “m1.small” i.e. small ec2 instance – performance sux, and it is actually cheaper to use a more expensive instance, “medium - cpu” or c1.medium.

http://diznix.com/dizwell/archives/153 Full Text Search benchmark, Oracle vs Postgres, on 2x Quad core Xeon, 8G RAM, 2 SSDs x RAID 1. Pg is being slightly faster when query contains several words, once word count increases, oracle leaps ahead (suggesting a more mature FTS implementation)

http://blogs.azulsystems.com/cliff/2009/09/java-vs-c-performance-again.html somewhat useful comparative review

http://thread.gmane.org/gmane.comp.file-systems.btrfs/4311/focus=4342 Currently ext2, will move to ext4 . Read the discussion for more info

http://www.randombugs.com/linux/mysql-postgresql-benchmarks.html yet another mysql vs postgresql benchmark using “OSDB”. Small difference between the two, but consider that benchmark was done on 2-way SMP, 3GB ram, 1 SATA DISK and a small dataset.

http://vegan.net/lb/archive/11-2009/0022.html discussion on LB-L (load balancing mailing list) about load-balancer performance benchmarks, vendor’s BS and so on.

http://blog.matt-darby.com/posts/697-apache-passenger-vs-nginx-mongrel comparing 2 popular ways to deploy ruby-on-rails app

http://vger.kernel.org/netconf2009_slides/nehalem_pps_comparison.pdf test results for linux routing performance on recent intel “nehalem” CPUs and 10G ethernet

http://zooie.wordpress.com/2009/07/06/a-comparison-of-open-source-search-engines-and-indexing-twitter/ this is not a strict benchmark, but rather a broad review. Results should be taken as a starting point in research, not as benchmark results.