Re: [Beowulf] Cloud / HPC

2013-04-16 Thread Mark Hahn
> http://www.admin-magazine.com/HPC/articles/the_cloud_s_role_in_hpc I had a very hard time with this article. in the "Massively Concurrent Runs" section, I really couldn't parse it as anything other than a devil's advocacy of people getting access to big clusters. it would be great if every r

Re: [Beowulf] Definition of HPC

2013-04-16 Thread James Cuff
> > "Owned compute" has some advantages over "rented compute." In general, the > > control one has over one's owned resources enables applications to run with > > greater performance. Some optimizations just demand root access! Yep, and then stuff like this happens http://blog.jcuff.net/201

Re: [Beowulf] Definition of HPC

2013-04-16 Thread Lux, Jim (337C)
On 4/16/13 12:50 PM, "Max R. Dechantsreiter" wrote: >...Perhaps HPC should be defined problem-down, rather than hardware-up. > >Regarding your "Issue #1:" > >There are some _nearly_ EP problems that require very minimal, but >crucial, >communication (perhaps at long intervals). > >As to your "I

Re: [Beowulf] Definition of HPC

2013-04-16 Thread Ellis H. Wilson III
On 04/16/2013 03:50 PM, Max R. Dechantsreiter wrote: > ...Perhaps HPC should be defined problem-down, rather than hardware-up. > > Regarding your "Issue #1:" > > There are some _nearly_ EP problems that require very minimal, but crucial, > communication (perhaps at long intervals). Fair enough, bu

Re: [Beowulf] Cloud / HPC

2013-04-16 Thread Douglas Eadline
>> >> Finally, I did just post and article on some experiments with small >> scale >> switchless 10GigE, http://goo.gl/5l2NJ if you are interested in that >> kind of thing. > > Very interesting. I realise this is not an RSS feed channel, but > can you perhaps post such notifications when interesti

Re: [Beowulf] Definition of HPC

2013-04-16 Thread Prentice Bisbal
On 04/16/2013 03:50 PM, Max R. Dechantsreiter wrote: > ...Perhaps HPC should be defined problem-down, rather than hardware-up. > > Regarding your "Issue #1:" > > There are some _nearly_ EP problems that require very minimal, but crucial, > communication (perhaps at long intervals). > > As to your "

Re: [Beowulf] Definition of HPC

2013-04-16 Thread Joe Landman
On 04/16/2013 03:50 PM, Max R. Dechantsreiter wrote: > ...Perhaps HPC should be defined problem-down, rather than hardware-up. > > Regarding your "Issue #1:" > > There are some _nearly_ EP problems that require very minimal, but crucial, > communication (perhaps at long intervals). > > As to your "

Re: [Beowulf] Definition of HPC

2013-04-16 Thread Max R. Dechantsreiter
...Perhaps HPC should be defined problem-down, rather than hardware-up. Regarding your "Issue #1:" There are some _nearly_ EP problems that require very minimal, but crucial, communication (perhaps at long intervals). As to your "Issue #2:" "Owned compute" has some advantages over "rented compu

Re: [Beowulf] Cloud / HPC

2013-04-16 Thread Hearns, John
> > Finally, I did just post and article on some experiments with small scale > switchless 10GigE, http://goo.gl/5l2NJ if you are interested in that > kind of thing. Very interesting. I realise this is not an RSS feed channel, but can you perhaps post such notifications when interesting articles

Re: [Beowulf] Cloud / HPC

2013-04-16 Thread Eugen Leitl
On Mon, Apr 15, 2013 at 01:20:26PM -0400, Douglas Eadline wrote: > If you want to help the community, you can write an article > for ClusterMonkey (contact me off list), visit the site once > and a while, or refer people to it when they are interested > in learning about cluster HPC. Posting links

Re: [Beowulf] Cloud / HPC

2013-04-16 Thread Hearns, John
> I've taken to saying I work in "Computing" as a distinct field from "IT". > The difference being that Computing is about using computers for > calculations/analytics rather than as a tool for doing something else. Ashley, you are hitting a nail on the head there. Throughout my career, I have