March 23, 2011
For those who are used to running heavy-duty workloads on dedicated machines, it might seem like a stretch to say that it’s possible to procure and provision an HPC environment in minutes. This has been the highly suspect but much discussed “miracle” of IaaS.
This week Amazon Web Services is asking potential users to suspend this disbelief via a trial credit and video tutorial that is meant to show that this process isn’t complex.
The video, which provides live screenshots of the processes behind creating an HPC environment, uses a molecular dynamics tool to demonstrate how the process works following initial setup. From the selection of an HPC compatible AMI, choosing the number of servers, defining placement groups (to ensure low-latency communication), creating credentials and security groups, the tutorial is wide in scope when one considers that it’s only a tick over ten minutes in length.
Aside from using the video and demo credits to actually run an application, this is a good intro for those who have been keeping up with all the talk about public cloud resources but haven’t taken them for a spin yet. It’s difficult to get a sense of the details behind using cloud resources if you’ve never broken the seal on an interface, even if it’s one you don’t plan on using.
One thing that the non-technically minded viewer who is just nosing around to see what it’s all about might notice, however, is that while the process is simple enough it does require some knowledge that might not be gleaned over the course of a ten-minute video.
For instance, while you might have an idea of what OS you want to remain glued to, what instance type will be best for your particular application? Of course you can experiment with this to some degree but the right (or wrong) choices can end up costing you in the end. Amazon provides a thorough overview of what the instances entail but from conversations with end users, both those with and without much experience using Amazon’s IaaS offerings, the choices are not always clear cut.
One of the hallmarks (or some might say weaknesses) of Amazon is that it’s an Infrasturcture-as-a-Service provider in the most stark possible way…In other words, you’re given access to the machines you need but beyond that there is not as much hand-holding as some need to get up and running. Videos like the one in question provide a good jumping off point for getting the environment spun up but there’s some background know-how that needs to be present.
At the very least, for those who are new to the public cloud user interface, this walkthrough gives a succinct sense of where to start as you firm up your background knowledge to select the right instance types and further refine and define your HPC environment.
Full story at Amazon Web Services
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?