January 31, 2001

LinuxWorld: Preview of dual-processor DDR Athlon Linux performance

Author: JT Smith

- By Jeff Field -

For years, dual processor x86 systems have only been possible with Intel CPUs, and for the most part on Intel chipsets. AMD has come up with an answer to this, its 760MP chipset. At LinuxWorld, I was given the opportunity to use a dual Athlon system with DDR memory on a prerelease 760MP board manufactured by Tyan, and initial impressions are very positive.

The disclaimer
The system used for this test was a preliminary system from ASL. The system was present in the MandrakeSoft booth with a rather small sign above it that said "World's First Dual DDR Athlon." This system is a prerelease motherboard running a prerelease chipset. As such, these results may not reflect the final performance of a similarly configured system. It is, however, a very good estimate.

The setup
The CPUs used in these tests were two AMD Athlon 1.2GHz CPUs. As far as I could tell, these were standard Athlon processors. The CPUs were running on a 266MHz frontside bus connected to 256 megabytes of PC2100 DDR RAM. The chipset was a prerelease 760MP running on a Tyan motherboard. The system was running Mandrake 7.2 with a 2.4 kernel. The kernel used in the kernel compiles was 2.4.0ac12.

The results
For this review, the only benchmarks available to me were 2.4.0ac12 compile times; however, they should more than suffice. The first kernel compile I did was a single processor compile of 2.4.0ac12. The kernel was configured with the default options for "make config." You can reproduce this fairly easily by typing "make config" and holding down the enter key for a while. The kernel was then compiled using "time make bzImage." The dual processor results were then done by first doing "make clean" then "time make -j3 bzImage".

Kernel Compile Times (Minutes:Seconds)

Single/Dual Athlon 1.2GHz
Single Dual
2.4.0ac12 4:51 2:00

While the results for the single-CPU compile were good results, what we are really looking at are the dual results. The kernel compile with the single processor was 142% slower than both processors, an impressive increase showing that adding another CPU doesn't necessarily mean just a 100% boost in performance. Although a two-minute 2.4.0ac12 compile is definately impressive.

As I stated before, this is by no means a production model system, but it is an excellent look at the potential performance of dual CPU Athlon-based systems. Having alternatives to Intel in the multi-processor x86 market can only help the end user, because competition will reduce prices, and Athlons are (for the most part) available for less than the equivilent Intel CPU. Once these systems become widely available, I wouldn't be surprised if AMD suddenly gained a large piece of the dual processor workstation and server market.

NewsForge editors read and respond to comments posted on our discussion page.


  • Unix
Click Here!