[Beowulf] HPC cloud bursting providers?

[Tim Cutts]

We don't actively use it, but we have demonstrated we can extend our LSF clusters into AWS.  I'm sure it could be done just as easily to other public cloud platforms.

It's a more practical prospect for us than for many traditional HPC sites, because most of our workloads are embarrassingly parallel single threaded tasks.

Having said that, many of our tasks require a lot of RAM, and public cloud VMs tend to get disproportionately expensive if you want a lot of RAM per thread.

Cloudbursting is also not particularly great if your datasets are large, since obviously you lose both time and money shipping the data in and out of the public cloud.

So technically it's fairly easy to do, but whether it's worthwhile is very workload-dependent.

Regards,

Tim

-- 
Head of Scientific Computing
Wellcome Trust Sanger Institute

On 21/02/2017, 00:03, "Beowulf on behalf of Jeff Friedman" <beowulf-bounces at beowulf.org on behalf of jeff.friedman at siliconmechanics.com> wrote:

    Hi all,
    
    Has anyone dealt with bursting from an on-prem HPC cluster to the cloud before? Are there any providers that stand out in this category? Quick searches reveal the usual suspects (AWS, Google, etc). Just wondered what real world experience has to say… :)
    
    Thanks!
    
    Jeff Friedman
    Sales Engineer
    c: 206.819.2824
    www.siliconmechanics.com
    
    
    _______________________________________________
    Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
    To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
    




-- 
 The Wellcome Trust Sanger Institute is operated by Genome Research 
 Limited, a charity registered in England with number 1021457 and a 
 company registered in England with number 2742969, whose registered 
 office is 215 Euston Road, London, NW1 2BE.