This past summer, University Research Computing and ITS teams worked to overhaul the HPC environment. As a result of extensive upgrades, users of high-performance computing at UNC Charlotte are experiencing faster computing and increased capacity for storage.
About the Upgrades
The HPC network environment is made up of two components: the network equipment (Cisco) and the InfiniBand (IB) equipment. Both are five years or older, and the Cisco equipment will reach end-of-support in December 2019. Both components were replaced, dramatically increasing both the speed and performance of the cluster. The network speed increased from 1Gbps/10Gbps to 40Gbps while the IB speed soared from 40Gbps to 100Gbps. This speed increase results in a corresponding decrease in job completion time as storage can be accessed quicker.
The HPC storage environment is made up of two storage systems: NFS and Lustre. Lustre is the primary storage system for the cluster, and NFS is the central storage system for project space, user home space, and those with many small files. The upgrade increased the speed and size of the Lustre environment, as well as increase the pace for the NFS environment. Lustre's capacity rose from 1PB to 2PB of total storage, while NFS speed increased from 10Gbps to 40Gbps.
Data Transfer DMZ
The previous HPC environment did not allow for incoming high-speed data transfer, which presented problems when researchers had large datasets that needed to be moved into the cluster. The upgrade included the creation of a high-speed data transfer demilitarized zone (DMZ) for the research cluster. The DMZ created a dedicated 10Gb data transfer channel (can be increased as needed) for researchers to bring data into the cluster.