Nor-Tech HPC Clusters Power Groundbreaking Projects

Print Friendly, PDF & Email

nor-techToday system integrator Nor-Tech disclosed that the company is working closely with some of the world’s top researchers and innovators to develop, build, deploy and support simulation clusters.

This has been an extremely exciting year for us that has allowed us to collaborate on innovations that promise to be groundbreaking and also discoveries that are changing the way we look at the universe,” said Nor-Tech President and CEO David Bollig.

The company’s most recent high profile project was building clusters for the LIGO Gravitational Wave team. One of the simulation clusters, nicknamed ORCA (Orange County Relativity Cluster for Astronomy) was instrumental in the first observation of a gravitational wave 10 months ago and a second wave that was recently detected. It was ORCA that provided confirmation that the first observed phenomenon was indeed a gravitational wave–the actual measurements fit the computer simulation exactly.

The cluster is housed at California State University, Fullerton in the University Data Center. It has more than 1500 gigabytes of memory, more than 30 terabytes of local storage, and 576 compute cores that together are capable of more than seven trillion operations per second.

Another ongoing project that has benefited from Nor-Tech’s HPC clusters is The IceCube Neutrino Observatory, which examines the cosmos from the vantage point of the South Pole. It is the world’s largest neutrino detector, with the goal of solving some of the most complex physics puzzles to date. About 300 physicists from 45 institutions in 12 countries make up the IceCube Collaboration. Using Nor-Tech’s clusters, researchers are currently in the process of analyzing and interpreting the data in order to help validate or invalidate theoretical models.

IceCube was one of the first projects to take advantage of Nor-Tech’s game-changing innovation that allows the substitution of consumer-grade GPUs for much more expensive commercial-grade GPUs. The technology community has long been aware that consumer-grade GPUs perform as well, and in some cases better than commercial grade GPUs at a price that is anywhere from 2/3 to 3/4 less. The main sticking point has been the traditional supercomputer chassis configuration, which only accommodates commercial grade GPUs.

Nor-Tech worked with sheet metal engineers to redesign the traditional supercomputer chassis so that it can accommodate consumer-grade GPUs. Then Nor-Tech engineers solved remaining engineering roadblocks.

Nor-Tech’s clusters are also proving instrumental in other projects that involve proprietary innovations. “We are a trusted partner with many of the most pioneering companies in the U.S. and beyond,” Bollig said. “They rely on us for smartly engineered, quality equipment that is easy to deploy. Another primary value we bring to the table is accessible support—meaning our engineers and the rest of our support team is readily available—we don’t put our clients on hold.”

Sign up for our insideHPC Newsletter