The UberCloud Experiment Shows Why #HPCMatters

This entry is part 2 of 7 in the series #HPCMatters

Every year when we gather for SC, we wait with bated breath for the announcement of the next Top500 list. It’s our own little celebration of the ever-growing size and power of the high-end supercomputers. We use these systems to answer some of the most complex questions out there.

However, bigger is not always better. It depends heavily on what one is trying to do and what resources are available. With costs continuing to spiral upwards for the top-end systems, there are fewer and fewer organizations that can afford to play within this elite group.

And, that’s okay. Not everyone really needs to have one of these systems. Seriously, if I had one, I’m not really sure what I’d do with it (other than to play the world’s largest game of Pong). I’m happier at a much smaller scale.

Because #HPCMatters there, as well.

And, just as there are many working on the problems on the top-end of the scale, there are many working on other sets of HPC-related issues. One of my favorite projects along these lines is The UberCloud Experiment (full disclosure: I am a registered participant of the project, so I may be a bit biased).

If you aren’t familiar with the project, it’s basically an experiment to understand how cloud can be better utilized for HPC work. From their website:

UberCloud [is] the free, community driven effort that explores the end-to-end processes of accessing and using remote computing resources in High Performance Computing (HPC) centers and tech computing clouds, on demand.

This is accomplished through a series of projects (now over 150), run by temporary project teams that bring together software, hardware, cloud and end-user experts dedicated to exploring a particular need. Their reports and webcasts are then made available to the community and the public.

We tend to celebrate the work done on the largest systems, but there’s amazing work being done throughout the entire community, big or small. The UberCloud Experiment is one of the best ways to see this.

Consider joining the experiment! It’s another great way to participate in the worldwide community and to see just how much #HPCMatters.

Series Navigation<< #HPCMatters – Growing Your FoodPrivate: TORQUE BoF 2014: Past, Present and Future >>
Facebook Twitter Email

Comments

  1. Trev, shame on me, but I discovered your nice blog just now, while searching for actual posts about our UberCloud Experiment, because we just announced the next Round. This cloud experiment has not lost it’s attraction; engineers are still joining and performing their individual cloud explorations, in a crowd-sourcing manner. As an example, the currently last experiment is currently led by Stanford University, with a Living Heart Project (LHP) experiment funded by HPE, about” Development and Calibration of Cardiac Simulator to Study Drug Toxicity”, and SIMULIA (Abaqus) as the software partner, and Advania as the cloud provider. Interested engineers can apply here: https://www.theubercloud.com/ubercloud-experiment-help/. Thanks.

Speak Your Mind

*