Project Grant 2350324

Award Date 6/15/24
Completion Date 5/31/27
Dollars Obligated $275K
Federal Grant Program
47.070
Assistance Type
Project Grant
Place of Performance
Charlotte, NC 28223, USA
Similar Awards
This $154,959 Project Grant awarded by the National Science Foundation (NSF) Integrative Activities (IA) program will fund research to develop efficient and intelligent Unified Memory (UM) design techniques for GPU-accelerated systems. The project, titled "COLLABORATIVE RESEARCH: SHF: SMALL: ACCESS PATTERNS IN MASSIVE PARALLELISM: ABSTRACTIONS AND ORIENTED UNIFIED MEMORY MANAGEMENT FOR GPU ACCELERATED SYSTEMS", aims to address the challenges of managing data migration between host...
This $182,564 Project Grant awarded by the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) supports research to develop scalable and reliable composable computer architectures in the post-Moore's Law era. The project aims to address challenges in managing integrated computing and memory components for efficient communication, and achieving sustainable performance on multi-GPU systems with disaggregated memories. The research will...
The National Science Foundation (NSF) awarded a $400,000 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to The University of Central Florida Board of Trustees Office of Research Division (UCF) to conduct research on taming large memory page challenges for memory bulk operations using a hardware/software co-design approach. The goal is to significantly improve the speed and bandwidth of single-level DRAM combined with ultra-low latency, non-volatile...
The National Science Foundation (NSF) awarded a $592,937 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to the University of Utah. The grant supports the development of cache-coherent accelerators to efficiently support persistent memory programming. The proposed approach aims to enable existing software to benefit from faster persistent memory storage without requiring changes to the code, while also providing safety during system crashes and...
This Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) provides $1,638,449 to develop the MemSysExplorer framework. This framework will provide the research community with tools to explore and evaluate next-generation memory systems across the application, system, and technology design spaces. The project, led by Tufts University with sub-awards to Harvard University and Amherst College, will create a...
The National Science Foundation (NSF) awarded a $215,908 Project Grant to the Regents of the University of California, Riverside under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070) to develop a comprehensive memory management ecosystem called MEMDRIVE. The project aims to optimize memory usage throughout the application and system stack for resource-constrained embedded systems, addressing the "memory wall" problem that limits performance and...
This National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) grant award, under CFDA 47.070, provides $292,237 to the University of Illinois to develop a framework that empowers non-computer science researchers to harness the power of modern hardware with massive parallelism for training advanced AI models for scientific discovery. The key products and services to be delivered include: Novel memory-efficient and hardware-friendly kernels that can capture...
The National Science Foundation (NSF) awarded a $221,138 Project Grant under the Computer and Information Science and Engineering (CFDA 47.070) program to the University of California, Merced. The grant, titled "REFORMING PROFILING TECHNIQUES TO GUIDE SYSTEMIC PERFORMANCE TUNING FOR GPU-ACCELERATED DEEP LEARNING WORKLOADS," aims to develop advanced profiling techniques to enable cross-layer performance optimization of deep learning models running on graphics processing units (GPUs)....
This $228,579 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program supports research by The Ohio State University to develop hardware and software solutions to address fundamental challenges in enabling in-memory computing for memory-intensive applications. The project takes a holistic approach, spanning multiple system layers, to tackle issues around fragmented memory resources, architectural support for memory sharing,...
This $1,091,988 federal Project Grant awarded by the National Science Foundation (NSF) under the Computer and Information Science and Engineering (CFDA 47.070) program supports the development of a new framework to systematically design and optimize high-performance graph analytics algorithms. The researchers at the University of California, Davis (UC Davis) will create an open-source software platform that allows for automated exploration of implementation choices for graph computations using a...

This $274,765 project grant awarded by the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) establishes a foundation for efficient and intelligent unified memory design to harness the power of advanced GPU accelerators. The key products and services to be delivered include:

  1. Developing an abstraction framework called ACCORD to capture the spatial and temporal patterns of massively parallel memory accesses, enabling quantitative assessment of their interaction with unified memory (UM) techniques and optimization of algorithms and UM systems.

  2. Designing access pattern-oriented UM techniques for online deployment to support various scientific and deep learning workloads, aiming to reduce time-to-production and time-to-completion for these applications.

  3. Integrating the ACCORD framework into real-world UM systems to improve the performance and scalability of GPU-accelerated computing.

The project also includes rich education, outreach, and broadening participation activities to recruit and support students from underrepresented groups in computer science and engineering.

Generated 3/18/25, 4:32 AM