Comments on: Early results from DOE cloud experiment: MPI applications are slower

By: Amazon adds support for traditional HPC workloads with Cluster Compute instance | insideHPC.com

Tue, 13 Jul 2010 14:08:07 +0000

[...] virtualized servers underperform on these types of applications (lots of writing on this, but see here and here for examples). Vogels acknowledges this in his post As much as Amazon EC2 and Elastic Map [...]

By: What to read this week | insideHPC.com

What to read this week | insideHPC.com — Fri, 25 Jun 2010 18:58:27 +0000

[...] Early results from DOE cloud experiment: MPI applications are slower [...]

By: John West

John West — Thu, 24 Jun 2010 13:30:27 +0000

Kathy – thanks so much for checking in and following up.

By: Kathy Yelick

Kathy Yelick — Thu, 24 Jun 2010 02:36:48 +0000

Since there has been some followup discussion, I wanted to clarify and add some context. The factor of 10 was a comparison between our unvirtualized Magellan hardware and Amazon’s Elastic Compute Cloud (EC2) using m1.xlarge instances. We ran the NERSC6 benchmarks to perform the comparison. For the seven applications we tested, the mean slowdown factor for EC2 relative to Magellan was 10.8. The best application, GAMESS, was 2.7 times slower, while the worst performance was with PARATEC, which was 51.8 times slower. Again, the Magellan results were on unvirtualized hardware with an Infiniband interconnect.

We are in the process of repeating the benchmarks on Magellan and varying aspects of the configuration to better isolate the reasons for the slowdown and to enable more cloud-like features on the Magellan testbed. The goal is to understand the value of clouds for scientific computing.

By: John West

John West — Tue, 22 Jun 2010 13:59:09 +0000

No, I didn’t miss that point, Greg. In fact, Ian and I had a long comment thread on my site and his at the time I pointed to Walker’s paper last year. In a private (and thus relatively finite-sized) cloud like NERSC’s a substantial move to a cloud platform for science would still have the wait times that a traditional batch system would have (more demand than capacity). Also, knowing that the cloud software/hardware stack itself is inherently slower puts a lower bound on run time.

By: Best Nine Brainiac’s Experiment in Microwaves | PCB Electronics Designing

Mon, 21 Jun 2010 19:44:24 +0000

[...] Early results from DOE cloud experiment: MPI applications are … [...]

By: Greg Pfister

Greg Pfister — Mon, 21 Jun 2010 18:47:43 +0000

You missed mentioning an important point raised by Ian Foster a while ago in his blog: The job run time may be slower, but the total time to solution much better because you don’t have to wait in a queue.

Or maybe he was subliminally pushing for more or bigger supercomputing centers to reduce the queue lengths.

By: -

Mon, 21 Jun 2010 18:09:44 +0000

This article would be much better if it told what are the slowdowns compared to.
Actually I thought it’s comparison against some other way of expressing parallelism and I had to read the original link to find out that I was wrong.