The National Science Foundation (NSF) awarded a $295,548 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to the University of Illinois. The award will support the development of an "Intelligent Management of Hybrid Workloads for Extreme Scale Computing" framework. Key research activities include modeling the performance implications of diverse workloads on supercomputers, creating new intelligent multi-resource scheduling methods, and...
The National Science Foundation (NSF) awarded a $174,200 Project Grant under its Computer and Information Science and Engineering (CFDA 47.070) program to the University of North Carolina at Chapel Hill (UNC-CH) to develop new resource allocation policies for optimizing the scheduling of parallelizable machine learning (ML) training workloads on shared hardware clusters. The goal is to enable the rapid and efficient training of highly accurate ML models using limited computing resources. This...
The National Science Foundation (NSF) awarded a $406,004 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to Carnegie Mellon University (CMU). The goal of this 4-year project is to develop new resource allocation policies that enable efficient and timely training of machine learning models by leveraging parallelizable computing resources. The research will focus on modeling the unique characteristics of machine learning training workloads, such as...
This $300,000 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) aims to develop strategies for efficiently leveraging idle resources in high-performance computing (HPC) systems to accelerate large-scale artificial intelligence (AI) workloads. The key objectives are to: 1) analyze patterns of idle resources in HPC environments, 2) develop methods to safely and rapidly harvest these idle resources, and...
This Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) provides $564,958 to Carnegie Mellon University from December 1, 2024 to November 30, 2027. The project aims to advance fundamental knowledge and innovate resource allocation algorithms to address the challenges posed by high variability and uncertainty in both demand and service in modern computing systems, especially those supporting machine learning...
This $275,000 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) aims to develop and test new scheduling algorithms for improving the performance of modern database systems. The goal is to create models and policies that can efficiently allocate limited hardware resources, such as compute and memory, to handle a stream of parallelizable database queries with varying levels of parallelizability and service...
This Project Grant award for $174,925, provided by the National Science Foundation (CFDA 47.070 - Computer and Information Science and Engineering), supports the development of novel data transfer optimization techniques for high-performance computing applications. The key project objectives are to: (i) implement a robust monitoring framework to enhance system visibility and enable global optimization and fair resource allocation; (ii) develop a simulator that replicates complex storage and...
This Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) will support research at the University of California, Santa Cruz (UCSC) to develop new methods for online scheduling algorithms. The $190,473 award, effective July 1, 2025 through June 30, 2026, aims to investigate the effectiveness of online scheduling techniques for a variety of computing problems. The research objectives include developing new...
This Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program aims to develop algorithms and theoretical analysis for large-scale resource allocation. The $129,536 project, awarded on April 15, 2025, will run through March 31, 2030. The research seeks to address three core challenges in resource allocation: (1) effective preference elicitation from users, (2) accounting for preference uncertainty, and (3) efficient...
This Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program, CFDA 47.070, provides $300,000 to the Trustees of the Stevens Institute of Technology to advance the efficiency and productivity of high-performance computing (HPC) systems by leveraging idle resources to expedite artificial intelligence (AI) workloads. The project centers on three interconnected research themes: 1) analyzing idle resources in HPC systems to...
This Project Grant award of $279,103 from the National Science Foundation's (NSF) Integrative Activities program (CFDA 47.083) supports the University of Kansas Center for Research Inc. in developing an integrated algorithmic and machine learning-based approach to enhance resource scheduling in high-performance computing (HPC) systems. The key goals are to address uncertainty in emerging HPC applications, resource contention, and workflow variations by designing novel scheduling algorithms and leveraging predictive machine learning models. The project aims to substantially improve the efficiency of HPC resource utilization across scientific domains like neuroscience, medical research, climate modeling, and artificial intelligence. Additionally, the award includes educational and outreach initiatives to engage students from K-12 through graduate levels, particularly targeting underrepresented and underserved communities in STEM and computer science. The project will run from October 2025 through September 2030.