While CentOS, Scientific Linux, and Oracle Linux Server are all derived from the same upstream source (Red Hat Enterprise Linux), how does the system performance compare between these RHEL derivatives? Here are some benchmarks of each of the 6.1 releases for Oracle Server, CentOS, and Scientific Linux, as they all do not perform the same.
After doing some Oracle Linux Server 6.1 testing over the weekend, I proceeded to install the recently released CentOS 6.1 and then Scientific Linux 6.1 to see how the performance of these Linux operating systems based upon Red Hat Enterprise Linux 6.1 compare in terms of performance. To some surprise, there are some noticeable performance differences in some tests.
All three RHEL derivatives were tested in their default configurations with the packages that ship with the OS installation by default. Being all based upon the same upstream source, this means that it is the Linux 2.6.32 kernel, Mesa 7.10, the GCC 4.4.5 compiler, and uses EXT4 by default as some of the key items worth noting. The 64-bit versions of Oracle Server 6.1, Scientific 6.1, and CentOS 6.1 were used.
Fedora 12/13 were also going to be tested since that's what Red Hat Enterprise Linux 6 is partially based upon, but those releases were not compatible with the hardware used in this article. If there is sufficient interest generated from this article, I will run more RHEL-based Linux distribution tests on a greater variety of hardware and more workstation/server-class hardware and tests.
Testing was done via the Phoronix Test Suite 3.6-Arendal release.