Time To Get To It: The Cluster Documentation Project
- Published on Monday, 14 May 2012 09:27
- Written by Administrator
- Hits: 111
Somebody better be writing this down for the next crew
Remember when you said, I'll get to the documentation later and never did. Well, now it is later and it is time to get the documentation in better shape. After contributing to on-line cluster documentation for years, Cluster Monkey has decided to start a community based project to help make something that is good much better.
The state of HPC and cluster documentation is not bad per-se, but it needs work. Most projects have adequate documentation for experts and many academic text books have good in-depth background. There seems to be a hole however, in the "how-to" type of cluster documentation. In particular, it is often hard for newcomers to the community/market to get a foot hold in the HPC thing as it were. The open source nature of much of the software has created a wide ranging spectrum of resources and critical pathways. The Cluster Documentation Project (CDP) is designed to help improve this state of affairs. Leveraging some past efforts, the current project site has begun, but is far from being complete. For instance it has a unique listing of Open/Freely Available Cluster Applications not found anywhere else.
Read more: Time To Get To It: The Cluster Documentation Project
Supercomputers: Getting it Right and Wrong
- Published on Wednesday, 28 March 2012 11:15
- Written by Douglas Eadline
- Hits: 839
Much has changed in the supercomputing arena. Even you can get in the game!
Recently, Sebastian Anthony wrote an article for ExtremeTech entitled What Can You Do With A Supercomputer? His conclusion was "not much" and for many people he is largely correct. However, there is deeper understanding that may change the answer to "plenty."
He was mostly right when talking about the worlds largest supercomputers. Indeed, one very workable past definition of a supercomputer was "any computer that had at least a six digit price tag." In the past, that was largely true and created a rather daunting barrier to entry for those who needed to crunch numbers. The cost was due to an architectural wall between supercomputers and the rest of computing. These systems were designed to perform math very quickly using vector processors. It all worked rather well until the cost of fabrication made creating your own vector CPU prohibitively expensive.
SC11: Intel MIC (Knights Ferry) Interview
- Published on Wednesday, 12 January 2011 19:00
- Written by Douglas Eadline
- Hits: 378
Back in November (yes, November) at SC11, I had a chance to sit down with Jim Ang, Technical Manager at Sandia National Laboratories in Appro's booth. Appro was gracious enough to allow me to interview Jim about his initial experiences with Intel's new Knights Ferry hardware and software. Similar to a GP-GPU from NVidia or AMD, Knights Ferry works in conjunction with a host server and is based on parallel x86 cores or the Intel MIC Architecture.
A Smidgen of Quantum Computing
- Published on Monday, 05 March 2012 11:22
- Written by Douglas Eadline
- Hits: 5625
Does a real instant password cracking quantum computer exist, or are we just at the beginning of the next revolution in computing?
(Note: The first half of this article was published last year at Linux-mag.com, however the second half was never published. Both parts have been updated and are presented below.)
Recent news by IBM describes very good progress toward the creation of real quantum computer. The work helped solve the stability problem associated with many quantum computer designs. There have been other quantum computer announcements over the last year. In May of 2011 a press release by DWave Systems proclaimed D-Wave Systems sells its first Quantum Computing System to Lockheed Martin Corporation. The press release created a bit of a stir in the quantum computing world and as with any new technology there is often some confusing (some deliberate) around actual milestones and press releases.
Welcome to the New Cluster Monkey
- Published on Wednesday, 29 February 2012 20:46
- Written by Douglas Eadline
- Hits: 342
Faster, better, cheaper, and hopefully not too broken
In case you have not noticed, we look different. There have been massive upgrades in the monkey kingdom. The front end is new and hopefully faster -- the template name was afterburner so it better be fast. And, because we are technical monkeys we kept the clean simple look with the focus on good content. To that end, we have tried to keep article presentation dense and clean (i.e. you can read more than two short paragraphs before hitting the next page button). The front page now has most of the widget type things on the left and content links on the right including the latest and popular articles links, a poll, and syndicated links to popular sites and blogs. As with our previous format, the stories are in the middle. We also are thankful to our sponsors who support us in keeping the banner advertisements to a minimum.
The backend server and delivery platform (Joomla) have been upgraded as well. Overall things should work much more smoothly and allow us to offer more features (see below). Unfortunately there were a few glitches with the upgrade. Content was automatically converted and it seems most of it came through just fine. We could not transfer the old comments, howerver. As time permits we will attempt to manually copy them from the old site. The polls also had problems and we need to write a script to convert to the current polling platform (or we may just archive some screen shots of the old poll results). Similarly a picture display module needs to be upgraded so we can recover some Beobash snapshots.
Overall, things went better than expected and the new hardware and software have put us in a good position for growth. Speaking of which, we are planing some new things. First, more content is in the works as well as some new features. Later this year we will be starting a monthly newsletter presenting technical webinars (free) to site subscribers (sign-up now on the left). We have a few other ideas we want to try out as well.
Our goal is to provide open quality cluster and High Performance Computing (HPC) information to our readers. We hope that the information generates discussion and conversation within the HPC community and enables users, administrators, and managers to benefit from the art and science of HPC clustering.
Page 1 of 4
«StartPrev Page1234- Next PageEnd»Search
Feedburner
Login Form
Who's Online
We have 66 guests and no members online
Latest Stories/News
Popular
Reader Poll
InsideHPC
-
Slidecast: Allinea Software – Meeting the Quest to Run Applications Faster
21 May 2012 | 7:00 am
-
Load Testing with the Cloud – This Week on inside* Publications
20 May 2012 | 3:42 pm
-
Student Cluster Competition Comes to ISC’12
20 May 2012 | 1:07 pm
Scalability Blog
-
… and it can talk …
18 May 2012 | 2:38 pm
-
Its … alive ….
17 May 2012 | 4:24 pm
-
After 4 years, our deskside JackRabbit unit decided to shrug off its mortal coil
16 May 2012 | 4:12 pm
Sicorp Cluster Blog
-
HPC Software -- ouch!
22 Feb 2012 | 1:50 pm
-
How Many Cores is "ManyCores"
31 Jan 2012 | 2:00 am
-
Microservers To The Rescue
19 Jan 2012 | 2:00 am
