Memory Optimization (MemOp)

Abstract

The Memory Optimization (MemOp) program is developing technology to optimize memory transactions in large scale computing systems. The demand for computing services is growing within both the U.S. Government and commercial industry. In response, new technical approaches are being developed to provide massive computation efficiently and cost effectively. In particular, distributed data centers with high-speed interconnects and customizable hardware, including graphics processing units (GPU) and field programmable gate arrays (FPGAs), are being used by service providers to achieve greater efficiency and improved processing performance. MemOp is exploring new memory architectures that more fully leverage emerging customizable hardware to deliver computing services reliably and at reduced cost. The more promising MemOp memory architectures will be implemented and evaluated in hardware and software. The technologies developed in MemOp will provide enhanced efficiency and improved performance for large scale computing systems.

Document Details

Document Type
Accomplishment
Publication Date
Oct 01, 2023
Source ID
9efe9f00b981ccf095b5e3cc5b6ca76b

Tags

Fields of Study

  • Computer science

Readers

  • Distributed Systems and Data Platform Development
  • Parallel and Distributed Computing.

Related Documents