ARC is happy to announce the release of a new cluster, named Cascades, available at cascades1.arc.vt.edu and cascades2.arc.vt.edu. Cascades is a 196-node system capable of tackling the full spectrum of computational workloads, from problems requiring hundreds of compute cores to data-intensive problems requiring large amount of memory and storage resources. Cascade contains three compute engines designed for distinct workloads:
- General – Distributed, scalable workloads. With Intel’s latest-generation Broadwell processors, 2 16-core processors and 128 GB of memory on each node, this 190-node compute engine is suitable for traditional HPC jobs and large codes using MPI.
- GPU – Data visualization and code acceleration! There are four nodes in this compute engine which have - two Nvidia K80 GPUs, 512 GB of memory, and one 2 TB NVMe PCIe flash card.
- Very Large Memory – Graph analytics and very large datasets. With 3TB (3072 gigabytes) of memory, four 18-core processors and 6 1.8TB direct attached SAS hard drives, 400 GB SAS SSD drive, and one 2 TB NVMe PCIe flash card , each of these two servers will enable analysis of large highly-connected datasets, in-memory database applications, and speedier solution of other large problems.
Cascades is one of the first deployments of Intel's Omni-Path network technology. Omni-Path is an interconnect designed specifically for HPC. A combination of routing and network flow optimizations allow for both high communication bandwidth and ultra-low latency on Cascades. Omni-path is at the leading edge of Intel’s development of ultra-high speed interconnect. As such, look for continued increases in performance as Intel releases new versions of MPI compilers to take advantage of this technology.
You may request a Cascades account here. To make it easier for research groups to get started, we have deposited 100,000 Cascades system units into each project with a nonzero balance on DragonsTooth or NewRiver. These SUs are good for 6 months.
Other questions or requests for accounts or allocations can be submitted using our usual forms.