DOE Patents title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Network-aware cache coherence protocol enhancement

Abstract

A non-uniform memory access system includes several nodes that each have one or more processors, caches, local main memory, and a local bus that connects a node's processor(s) to its memory. The nodes are coupled to one another over a collection of point-to-point interconnects, thereby permitting processors in one node to access data stored in another node. Memory access time for remote memory takes longer than local memory because remote memory accesses have to travel across a communications network to arrive at the requesting processor. In some embodiments, inter-cache and main-memory-to-cache latencies are measured to determine whether it would be more efficient to satisfy memory access requests using cached copies stored in caches of owning nodes or from main memory of home nodes.

Inventors:
;
Issue Date:
Research Org.:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1576397
Patent Number(s):
10402327
Application Number:
15/358,318
Assignee:
Advanced Micro Devices, Inc. (Santa Clara, CA)
Patent Classifications (CPCs):
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
DOE Contract Number:  
AC52-07NA27344; B608045
Resource Type:
Patent
Resource Relation:
Patent File Date: 2016 Nov 22
Country of Publication:
United States
Language:
English

Citation Formats

Roberts, David A., and Fatehi, Ehsan. Network-aware cache coherence protocol enhancement. United States: N. p., 2019. Web.
Roberts, David A., & Fatehi, Ehsan. Network-aware cache coherence protocol enhancement. United States.
Roberts, David A., and Fatehi, Ehsan. Tue . "Network-aware cache coherence protocol enhancement". United States. https://www.osti.gov/servlets/purl/1576397.
@article{osti_1576397,
title = {Network-aware cache coherence protocol enhancement},
author = {Roberts, David A. and Fatehi, Ehsan},
abstractNote = {A non-uniform memory access system includes several nodes that each have one or more processors, caches, local main memory, and a local bus that connects a node's processor(s) to its memory. The nodes are coupled to one another over a collection of point-to-point interconnects, thereby permitting processors in one node to access data stored in another node. Memory access time for remote memory takes longer than local memory because remote memory accesses have to travel across a communications network to arrive at the requesting processor. In some embodiments, inter-cache and main-memory-to-cache latencies are measured to determine whether it would be more efficient to satisfy memory access requests using cached copies stored in caches of owning nodes or from main memory of home nodes.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Sep 03 00:00:00 EDT 2019},
month = {Tue Sep 03 00:00:00 EDT 2019}
}

Works referenced in this record:

System and method for performing a speculative cache fill
patent, August 2004


Speculative Directory Lookup for Sharing Classification
patent-application, November 2007


Proactive Cache Coherence
patent-application, June 2018


Numa System Resource Descriptors Including Performance Characteristics
patent-application, October 2008


Non-Uniform Memory Access Latency Adaptations to Achieve Bandwidth Quality of Service
patent-application, June 2017


Methods and Apparatus for Content Delivery and Replacement in a Network
patent-application, September 2008


Maintaining Cache Coherence in a Multi-Node, Symmetric Multiprocessing Computer
patent-application, December 2011


Fishing Lure
patent-application, June 2017


System and Method for Intelligent Software-Controlled Cache Injection
patent-application, August 2008


Cache Coherency Using Die-Stacked Memory Device with Logic Die
patent-application, June 2014


Speculative Region: Hardware Support for Selective Transactional Memory Access Annotation Using Instruction Prefix
patent-application, August 2010


Memory System
patent-application, April 2017


Memory Systems and Methods Including Training, Data Organizing, and/or Shadowing
patent-application, July 2017


Prefetching of Data and Instructions in a Data Processing Apparatus
patent-application, September 2014