Low Cost/Low Power HPC

Print Friendly, PDF & Email

Our buddy Dr. Doug Eadline posted a cool article this week outlining what may become an interesting niche of HPC.  I’ll go out on a limb and say that everyone involved with HPC facilities, infrastructure or procurement has some how had to deal with the relative power threshold of a potential platform.  I’ve actually heard lab managers say to vendors: “budget is not the problem, I just can’t {power,cool} that much gear!?”

Doug picked up an article from Extremetech that stated that Intel Atom processors were actually more efficient from a cost/Thermal Design Power [TDP] standpoint than their big brother Xeon variants for certain search applications.  OK, I’ll bite, now lets look at the numbers.

I went to Intel’s website and got most of the information for both an Atom D510 (dual core, 1.66 GHz) and Xeon 5570 (quad-core, 2.93 GHz). Both of these represent currently available parts and I used Intel’s stated price for quantities of 1000. Next, I tried to find floating point results for both processors. There are no SPEC numbers for the Atom, but I did find some POV numbers for both the D510 and an i7 870 running at 2.93GHz. (POV is a rendering program that uses heavy floating point processing). Since the i7 and Xeon are very similar, it seems reasonable to use the POV number as a measure of performance (remembering it is a single measurement). The results of my comparison are in the Table below. Note for the TDP/Perf and Price/Perf ratios, lower is better.

Eadline’s finding are quite interesting.  From his results, he find that the Nehalem Xeon runs 1.8 times faster, generates 7.3 times as much heat and costs 22 times as much as the D510 Atom.

The Atom has a clear price-to-performance advantage over the Xeon, however, it is not quite that simple. Since you need at least 8 Atom processors to do the work of one Xeon, you need 8 separate nodes. The cost per Atom node includes things like motherboards, memory, etc. so the 3X price to performance advantage may quickly evaporate. It may be possible to narrow the performance gap or there may be better performing floating point applications suited for the Atom, but for now we’ll use this data as a first approximation. Also note, I have not considered the fact that with 8 Atom nodes, there is 8 MB of total distributed cache available to a parallel application.

This is quite a find.  There are quite a number of up-and-coming HPC niche applications that don’t utilize raw floating point arithmetic.  For these applications where floating point is not the cat’s meow and you have enough memory bandwidth available to service enough requests, the Atom platform might very reasonable sense.  For those interested in some of the these fields, I suggest you take a look at Dr. Eadline’s article.  You can find it here, at Linux Magazine.

Trackbacks

  1. […] our post from last week on Eadline’s musing about Atom clusters?  Well, this week we have a press […]