Bench Press

What do the LHC and cloud computing have to do with each other? Read and find out.

Not two phrases that generally go with one another. However, with the high demand of calculations required to do large ion simulations, it seemed only natural that CERN physicists in Switzerland start employing the benefits of cloud computing. Before using cloud computing, CERN physicists relied on its own pool of distributed computing which was managed by a scheduler called AliEn.

However, the need for more resources suggested a push towards cloud computing and inspired the birth of the Nimbus Project. Sponsored by Google’s Summer of Code, the Nimbus Project worked on providing integration between cloud computing’s dynamically allocated resources along with the already existing infrastructure at CERN. Thus, to tackle this problem, the Nimbus Project needed to create a virtual machine capable of supporting CERN’s heavy ion calculations to deploy on computers outside of their distributed computing pool. Luckily, the CernVM technology already had existing support for supplying “portable development environments” in case scientists needed to crunch data on their personal laptops and desktops.

However, the snag with this approach was that these virtual machines still needed to be integrated with AliEn for it to be of any use. In order to do that, CERN scientists needed computers in the cloud to have context-specific information involving certain configurations and security. But thanks to the Nimbus Context Broker, software created by the Department of Energy’s Argonne National Laboratories and the University of Chicago, this context-specific information was able to be provided securely without much overhead. In the end, the endeavor was a success and support for Xen images used by Amazon’s EC2 and Science Cloud’s has already been added.

“Commercial cloud providers such as EC2 allow users to deploy groups of unconnected virtual machines, whereas scientists typically need a ready-to-use cluster whose nodes share a common configuration and security context. The Nimbus Context Broker bridges that gap,” said Kate Keahey, a computer scientist at Argonne and head of the Nimbus project.

For me, this breakthrough is another testament to how new technologies are being used in different fields to further research and science. Modeling and cloud computing, both topics we’ve discussed before on Benchpress, are paving the way for the future and illustrates technology’s wide-reaching benefits to all aspects of science.