Loading…
Loading grant details…
| Funder | National Science Foundation (US) |
|---|---|
| Recipient Organization | University of Massachusetts Amherst |
| Country | United States |
| Start Date | Nov 15, 2024 |
| End Date | Oct 31, 2026 |
| Duration | 715 days |
| Number of Grantees | 3 |
| Roles | Principal Investigator; Co-Principal Investigator |
| Data Source | National Science Foundation (US) |
| Grant ID | 2430001 |
Harmony, a high-capacity and green research computing cluster, adds new functionality and capacity to the regional Unity Research Computing Platform. Harmony is designed to efficiently serve large numbers of researchers and students, expand equitable computing access, support the need for computing resources in educational settings, and provide a blueprint for other regional centers to deploy high-capacity, low-cost, and low-power consumption systems.
As a regional resource, Harmony provides research and education computing infrastructure to numerous institutions in New England, including University of Massachusetts and University of Rhode Island, as well as small and primarily undergraduate-serving institutions, such as Smith College, Mount Holyoke College, Bryant University, UMass Dartmouth, UMass Boston, and UMass Lowell. In addition, Harmony contributes both CPU and GPU compute resources to the OSG’s Open Science Pool for use by researchers and educators nationwide.
Harmony specifically targets workloads that require large ensembles of small jobs, interactive jobs, or persistent services. These workload archetypes are critical to a broad range of science research domains and prevalent in education settings. Harmony pairs efficient and high-capacity hardware with a flexible scheduler designed to maximize resource use.
Harmony features high core-count ARM architecture CPUs, which provide exceptional capacity and comparable performance to x86_64 CPUs at minimal cost and power consumption, and NVIDIA A16 and L4 GPUs for cost effective and power efficient GPU acceleration. Harmony uses the Nomad scheduler and orchestrator for resource management. Nomad enables workloads requiring persistent services, supports educational uses through strong integration of environment and workload distribution, and maximizes hardware utilization across the entire Unity ecosystem.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
University of Massachusetts Amherst
Complete our application form to express your interest and we'll guide you through the process.
Apply for This Grant