November 02, 2012
Over the last four months, the Uber-Cloud Experiment has developed into a thriving community of fellow colleagues who are increasingly benefiting from the use of remote HPC resources, including HPC in the cloud. With a successful Round One having just wrapped up, the organizers of the experiment, Wolfgang Gentzsch and Burak Yenier, have opened up participation for Round Two, which will officially commence at SC12, in Salt Lake City, on November 15. Want to know more? Keep reading...
How do the members of this community benefit?
They benefit by joining the free Uber-Cloud Experiment, and exploring together the end-to-end process of accessing and using remote computing resources for HPC applications. Today, this community consists of almost 200 organizations and individuals, everybody with the vision of enhancing their current computing capacity with powerful remote resources, on demand, at their fingertips. Gone is the headache when computing resources were scarce, simulation models didn't fit into memory, and computing took too long.
What is the Uber-Cloud HPC Experiment?
Round 1 of the experiment was mainly about a very first exploration of accessing and using compute resources remotely, hands on. And for many participants this was the first time ever that they gained access to remote computing resources. With minimal intervention into the process, we monitored each of the 25 teams and discovered the real roadblocks and how the teams have removed them (or not). We will soon publish our report addressing these findings.
You can see some of our current Round 1 participants here. In Round 1 they formed teams like Anchor Bolt, Resonance, Radiofrequency, Supersonic, Liquid-Gas, Wing-Flow, Ship-Hull, Cement-Flows, Sprinkler, Space Capsule, Car Acoustics, Dosimetry, Weathermen, Wind Turbine, Combustion, Blood Flow, ChinaCFD, Gas Bubbles, Side impact, and ColombiaBio. Want to read more about Round 1? Please see our first call for participation.
How does the experiment work?
Suppose the industry end-user is in need of additional compute resources, say for speeding up the design cycle, for simulating a more sophisticated geometry or complex physics, or for running many more simulations for a higher quality result. We, the experiment orchestrators, will jointly look at this end-user's application and requirements, select appropriate resources, software, and the best-suited HPC experts in our community. This 'Team of Four', the end-user, software provider, resource provider, and HPC expert will then implement and run the end-user's task and the results will be delivered back to the end-user. Finally, the whole team will extract lessons learned, and present further recommendations as input, which can be published as a case study.
Experiment Round 2 starts now
Now Round 2 will be quite different: more advanced, more professional, semi-automatic, with more participants from CAE and life sciences, more teams, closer to reality, with a commercial production angle, using tools for project management, and tool-based measuring of effort and cost. One of the highlights of Round 2 will be the Uber-Cloud Services Directory, where hardware, software, and expertise providers can advertise their services to our Uber-Cloud community and to the wider HPC and Digital Manufacturing community.
The Experiment Kick-off at SC12
The final webinar of Round 1 and the kick-off of Round 2 will take place in the Intel booth at the Supercomputing Conference (SC12) in Salt Lake City at 11:00 am (local time) on November 15. If you can't make it to SC12, the webinar will be aired live for our registered experiment attendees and as always, the slides will be made available to our registered experiment participants following the webinar.
The Experiment Managers
Wolfgang Gentzsch and Burak Yenier are the creators and facilitators of the Uber-Cloud Experiment. Wolfgang is an HPC veteran. Having worked in leading positions in research, academia and industry for some 30 years, Wolfgang is now an HPC consultant and the chairman of the ISC Cloud conference series for HPC and Big Data in the Cloud. Burak is the vice president of operations at CashEdge (now part of Fiserv), a software-as-a-service company in Silicon Valley, which provides innovative payments and aggregation solutions to financial institutions.
The ever-growing complexity of scientific and engineering problems continues to pose new computational challenges. Thus, we present a novel federation model that enables end-users with the ability to aggregate heterogeneous resource scale problems. The feasibility of this federation model has been proven, in the context of the UberCloud HPC Experiment, by gathering the most comprehensive information to date on the effects of pillars on microfluid channel flow.
Large-scale, worldwide scientific initiatives rely on some cloud-based system to both coordinate efforts and manage computational efforts at peak times that cannot be contained within the combined in-house HPC resources. Last week at Google I/O, Brookhaven National Lab’s Sergey Panitkin discussed the role of the Google Compute Engine in providing computational support to ATLAS, a detector of high-energy particles at the Large Hadron Collider (LHC).
Frank Ding, engineering analysis & technical computing manager at Simpson Strong-Tie, discussed the advantages of utilizing the cloud for occasional scientific computing, identified the obstacles to doing so, and proposed workarounds to some of those obstacles.
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.