Linux Performance: Different Distributions, Very Different Results

"The performance results of different distributions, even ones running the same kernel version, the same core libraries, and the same filesystem can be very, very different.

"I see this issue debated on countless Linux forums often without a lot of facts. "It's faster for me" won't convince anyone and rightly so. In a discussion on LXer.com a user named herzeleid asked exactly the right question: "I wonder why that is?" This little article grew out of my response.

"Different distributions are better suited to different hardware. The most obvious example of this, both on the home desktop and in the corporate server room, are differences in processor architecture. For most desktop users this boils down to whether you are running a 32-bit CPU or a 64-bit CPU. (Dual and quad core machines are generally multiple 64-bit CPUs nowadays.) Yes, a 32-bit distribution will run just fine on a 64-bit machine and for most ordinary tasks there really won't be much if any difference in performance. For CPU intensive tasks that take full advantage of a 64-bit processor a 32-bit OS will not perform well."

Complete Story

Related Stories:

• Oops! Removed all kernels! (Mar 08, 2009)
• Browser benchmarks 2: even Wine beats Linux Firefox(Feb 13, 2009)
• Linux 2.6.28 Kernel Benchmarks(Dec 29, 2008)
• Intel Linux Graphics Performance Q4'08(Dec 23, 2008)
• Java Performance: Ubuntu Linux vs. Windows Vista(Dec 19, 2008)
• Real World Benchmarks Of The EXT4 File-System(Dec 03, 2008)
• Fedora 10 vs. Ubuntu 8.10 Benchmarks(Nov 29, 2008)