CC* Compute: Accelerating Computational Research for Engineering and Science (ACRES) at Clarkson University, A Campus Cluster Proposal

克拉克森大学正在建设一个计算集群（ACRES：加速工程和科学计算研究），以支持与克拉克森的四个跨学科研究主题相一致的数据和计算密集型项目：数据分析、健康世界解决方案、先进材料开发和下一代医疗保健。ACRES促进了需要访问高性能计算（HPC）资源的高影响力合作研究的开展，使目前不实用/可行的研究成为可能，并且还通过学分课程，本科研究和现有的NSF REU网站支持学生学习机会，重点是HPC。作为一项校园资源，根据为确保公平使用而实施的排队政策，大学的任何教师或学生都可以使用ACRES。此外，ACRES还支持克拉克森加大对计算研究的关注，并集群聘用了计算活跃的教师。ACRES计算集群取代了现有的一个已有5年历史的高性能计算集群，该集群的计算能力为105万核/小时/年。对计算能力的研究需求已增长到确定的总计850万核心/小时/年。ACRES的大小可以满足当前需求和适度的近期增长，未使用的计算能力通过开放科学网格（OSG）共享，以使更广泛的科学界受益。这个新的计算资源通过1120核、高速ib互连、4个NVIDIA Tesla V100 gpu和40tb的刮刮存储提供9800核/小时/年。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Clarkson University is building a computational cluster (ACRES: Accelerating Computation Research for Engineering and Science) to support data and computationally intensive projects aligned with Clarkson's four interdisciplinary research themes: Data Analytics, Healthy World Solutions, Advanced Materials Development, and Next Generation Healthcare. ACRES facilitates the conduct of high-impact, collaborative research that requires access to high-performance computing (HPC) resources, enables research currently not practical/feasible, and also supports student-learning opportunities through credit-bearing courses, undergraduate research, and an existing NSF REU site focusing on HPC. As a campus resource, ACRES is made available to any faculty member or student at the University according to queueing policies implemented to ensure fair-access. And, ACRES supports Clarkson's increased focus on computational research and a cluster hire of computationally active faculty. The ACRES compute cluster replaces an existing, five-year-old high-performance compute cluster whose computational capacity provided 1.05M core-h/yr. Research need for computational capacity has grown to an identified total of 8.5M core-h/yr. ACRES is sized to meet current demands and modest near-term growth with unused computational capacity being shared via the Open Science Grid (OSG) to benefit the broader scientific community. This new computational resource provides 9.8M core-h/year through 1120 cores, high-speed Infiniband interconnect, four NVIDIA Tesla V100 GPUs, and 40 TB of scratch storage.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.