November 03, 2010
At Cloud Expo this week I spent a few moments talking with Reid Smith, who discussed some issues relevant to high-performance computing and network capabilities in the context of broader cloud services.
More specifically, we talked about the value of cloud services that are cognizant of the power of the underlying network. For those in the world of HPC this is the last thing you’d forget; that performance is, after all, a key piece of high-performance computing. But when reaching out to enterprise markets outside of low-latency-grabbing financial services, the networking message bears some repeating.
According to Smith, as they have seen with many of their financial services clients, there is an equal need for high levels of security as well as ultra-fast response times. The company has, as Reid states, carried the lessons over from these industry experiences into high-performance computing since they have customers with large datasets that need to be quickly moved back and forth from what was once the grid (now the cloud) and for such users, data portability is a key issue.
Reid Smith made a couple of good points, including his statement that what everyone seems to miss in conversations about cloud computing is that the network is where the cloud is delivered. Many people in HPC don’t usually forget this critical element, but for the enterprise users that were walking around this message might have granted some pause.
Another statement Reid made, which although general, struck me as being one of those “obvious but worth repeating” messages is that the cloud is simply another way of providing compute power. There’s something about the clarity of such a remark, perhaps because it removes the cloud from all the buzz word madness around it, that makes it worth remembering. In other words, despite what seems to be incredible confusion about clouds and the need for constant debates about what the term means, how it’s defined, and how the many solutions to make it more powerful and intuitive work, let’s just remember…this is simply another way of delivering the compute capacity researchers and enterprise users need.
Savvis is one of those companies that doesn’t appear as often as others in the more niche arena of HPC cloud computing news rounds, despite the fact that they do work in financial services and other HPC-oriented verticals.
I might not read much about them on a daily basis in the HPC cloud rounds, but they certainly had a presence here at Cloud Expo this week in Santa Clara with one of the busier booths of the afternoon—and they didn’t appear to be giving away anything of value either like some of the other vendors.
Not that a cloud-shaped air freshener is valuable necessarily but I never know who I’ll end up sitting next to on the plane back.
Posted by Nicole Hemsoth - November 03, 2010 @ 4:18 PM, Pacific Daylight Time
Nicole Hemsoth is the managing editor of HPC in the Cloud and will discuss a range of overarching issues related to HPC-specific cloud topics in posts.
No Recent Blog Comments
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?