The University of Bath is in the process of upgrading its HPC infrastructure, which it says is “supporting a growing and broad range of research activities across the university.” The previous system, Balena, will be replaced by a new Microsoft Azure-based cloud HPC environment called Janus. Janus is complemented by a “tailor-made on-campus” [high-throughput computing] cluster of software applications that cannot be moved from campus.”
Janus, the university said, makes it “the first university in the world to move nearly all of its HPC research to the Microsoft Azure platform”[.]The new cloud environment contains an unspecified number of HB-series, HBv2-series, HBv3-series nodes, which contain between 16 and 120 AMD CPUs per node; HC series nodes, which contain 44 Intel CPUs per node; Fsv2-series nodes, which contain between two and 72 Intel CPUs per node; and NCv3-series and NDv2-series nodes, which contain between one and eight Nvidia Tesla V100 GPUs per node.
These nodes are supported by a total of 34 TB of dedicated storage (including 16 TB of data storage and 16 TB of scratch space) and Mellanox EDR InfiniBand networks, all of which are VPN-connected to Bath over a 1 GB/s. Janus uses Azure CycleCloud for automated environment configuration and management and Slurm for cluster task scheduling.
Meanwhile, the new HTC cluster, called Anatra, includes eight compute nodes, each equipped with dual AMD Milan CPUs, 256GB of memory and 960GB of NVMe storage. Anatra is based on RedHat Enterprise Linux v8 and uses a Slurm scheduler.
The university says Janus will provide many benefits to its students and researchers, including “significantly shorter” computation times and “a wide variety of advanced computational options, which are regularly updated to meet evolving research needs.” Indeed, Janus’ announcement follows the university’s two-year exploration of cloud infrastructure suitability, including an HPC cloud services pilot project conducted last summer with two dozen researchers.
“We used Azure as part of the pilot project to run our computational fluid dynamics simulations with the OpenFoam program,” said Katharine Fraser, lecturer in the Department of Mechanical Engineering at the University of Bath. “It was great to be able to run these large simulations exactly when we wanted, instead of queuing, and Balena’s transition was easy with Azure with a user-friendly interface. … In the future, the wide variety of nodes will be beneficial for performing different types of simulations. The flexibility that cloud computing promises for technical research is really exciting!”
Both Janus and Anatra are still in development and “will gradually be made available to researchers and postgraduate students at the university.” The university is accepting applications for early access user groups.
Balena, who will eventually replace Janus and Anatra, is pictured in the header image and includes 196 Intel Ivy Bridge nodes and 17 Intel Skylake nodes, delivering 63.2 peak teraflops of computing power.