October 03, 2011
Most conferences provide an opportunity for event sponsors to get their messages across to attendees in one way or another, at the very least by providing a platform to talk amidst the glow of a PowerPoint presentation. Oftentimes, these overviews address audiences at large—and avoid the “big questions” about potential problems, drawbacks or other points of weakness.
At last week’s ISC Cloud event in Mannheim, Germany, twelve vendors who might otherwise have given long detailed presentations were instead corralled to present a rapid-fire two-slide overview of their relevance to the HPC cloud space, then were put on the hot seat to answer some of the most complex questions floating around about HPC clouds.
This wasn’t your standard “vendor hot seat” session; the two teams, each with six vendors were given questions and had a short amount of time to answer. Three expert judges from different perspectives (press, academic, and industry) then evaluated the responses, awarding points to the most successful, comprehensive answer.
Representatives from Intel, HP, SGI, Mellanox, QLogic, Microsoft, Gompute, Microsoft, Penguin Computing, Bull, Eucalyptus, Univa and Samsung were divided into teams—then led into questions by the event’s host, Addison Snell of Intersect360 Research.
For those without time to watch the highlights from the vendor showdown, there are a few key phrases that emerged from nearly all the vendors present. The overwhelming emphasis from each of the vendors is best described as making new strides to simplify access, create openness and portability, and refine critical hardware and software elements that make clouds possible for a user group with complex needs. This last element means everything from solving software licensing problems to creating hyper-efficient cloud hardware that provides the performance needed for HPC applications while recognizing the power consumption challenges of cloud providers.
There was definite value in hearing answers about the stickiest issues in high performance computing for clouds from the specialist vendors who cater to different regions of the cloud ecosystem. For instance, pitting Mellanox and QLogic against one another on the question of what is needed on the interconnect side for robust HPC clouds or asking system vendors like SGI how clusters need to change to meet these specific needs creates the opportunity for direct answers to questions that users are asking most often.
HP and Intel, for instance discussed the intersection between HPC and clouds, with HP’s Frank Baetke claiming that HPC has always revolved around issues of access. He says that cloud provides this access and in this lies the potential of cloud in coming years. From the user perspective, Intel’s Stephan Gillich pointed to access models that are based on the needs of HPC users with applications, interfaces and environment tailored to ease of use. He says that the benefits can’t be realized without careful selection of the underlying technologies that simplify access since that is the important goal.
Other compelling questions put to the panel revolved around the evolution of on-demand computing and the role of the hardware backbone for cloud providers hoping to draw in HPC customers. SGI, Penguin, Bull, and Gompute, all of whom have on-demand HPC offerings noted that the evolution of cloud computing has been swift, but for users, creating interfaces that are open, easily accessible, and tailored to the needs of HPC applications and users remain critical challenges. While that might sound like the general answer for such a question, the back and forth answers made vendors challenge one another on such generalities. If you have time to watch the panel, a great example of this is embodied in the exchanges between Mellanox’s Gilad Shainer and Alastair McKeeman from QLogic as well as the ping pong of answers between SGI’s Tanasescu and Bull’s Olivier David.
While the video provides a sense of the action by capturing highlights from the hour and a half showdown, it is difficult to capture the audience’s take on the material on film. Needless to say, some of the questions definitely provoked strong conversation between voting rounds from the galley—and definitely showed the vendors that these are questions that matter.
We have a great deal more coverage from last week's event coming during the course of this week, but this panel elicited a strong response from the attendees and provided a rich sense of the directions a number of vendors in this small community are heading.
May 23, 2013 |
he study of climate change is one of those scientific problems where it is almost essential to model the entire Earth to attain accurate results and make worthwhile predictions. In an attempt to make climate science more accessible to smaller research facilities, NASA introduced what they call ‘Climate in a Box,’ a system they note acts as a desktop supercomputer.
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 10, 2013 |
Australian visual effects company, Animal Logic, is considering a move to the public cloud.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/02/2012 | AMD | Developers today are just beginning to explore the potential of heterogeneous computing, but the potential for this new paradigm is huge. This brief article reviews how the technology might impact a range of application development areas, including client experiences and cloud-based data management. As platforms like OpenCL continue to evolve, the benefits of heterogeneous computing will become even more accessible. Use this quick article to jump-start your own thinking on heterogeneous computing.