[Beowulf] Cloud / HPC
john.hearns at mclaren.com
Mon Apr 15 02:17:13 PDT 2013
On 04/14/2013 12:46 PM, James Cuff wrote:
> Finally a well balanced,hype free article by respected industry expert
> Jeff Layton:
Just read it. This is a very good article.
"I consider cloud computing a tool or technique for solving research
computing problems. Nothing more or less. It's not a panacea, not should
it be ignored. Issues that must be addressed include data movement and
security, but it also can save you money and make your traditional HPC
resources stretch further. "
I agree. Very well put.
I've been making the argument for a while that what is euphemistically
called "Big Data" is an applied version of high performance computing
... not in a traditional sense (e.g. MPI, ultra low latency
interconnects, etc.), but focused upon analyzing huge swaths of data,
and turning it from "data" with noise into usable actionable
Again, I agree.
Coming from a high energy physics background, thats what experimental HPC folks have been doing for many years.
Huge datasets, with event parallelism and no real need for massively parallel computation.
(indeed remember that the CrayXMP-24 at CERN ended up being used as a hugge disk server!)
HPE is a statistical science - unlike thepress release image of a 'eureka' science, where a Higgs Boson is tracked down and found,
it is a matter of analysing mountaisn of data, and making sure that the statistical 'cuts' you apply are isolating genuine events above the background
The contents of this email are confidential and for the exclusive use of the intended recipient. If you receive this email in error you should not copy it, retransmit it, use it or disclose its contents but should return it to the sender immediately and delete your copy.
More information about the Beowulf