Balancing Performance, Area, and Power in an On-Chip Network

Several trends can be observed in modern microprocessor design. Architectures have become increasingly complex while design time continues to dwindle. As feature sizes shrink, wire resistance and delay increase, limiting architects from scaling designs centered around a single thread of execution....

Full description

Bibliographic Details
Main Author: Gold, Brian
Other Authors: Electrical and Computer Engineering
Format: Others
Published: Virginia Tech 2014
Subjects:
Online Access:http://hdl.handle.net/10919/34137
http://scholar.lib.vt.edu/theses/available/etd-07242003-134147/
id ndltd-VTETD-oai-vtechworks.lib.vt.edu-10919-34137
record_format oai_dc
spelling ndltd-VTETD-oai-vtechworks.lib.vt.edu-10919-341372020-09-26T05:38:52Z Balancing Performance, Area, and Power in an On-Chip Network Gold, Brian Electrical and Computer Engineering Baker, James M. Jr. Michael S, Hsiao Jones, Mark T. area virtual channels SCMP power network router crossbar switch single chip computer message passing system on chip Several trends can be observed in modern microprocessor design. Architectures have become increasingly complex while design time continues to dwindle. As feature sizes shrink, wire resistance and delay increase, limiting architects from scaling designs centered around a single thread of execution. Where previous decades have focused on exploiting instruction-level parallelism, emerging applications such as streaming media and on-line transaction processing have shown greater thread-level parallelism. Finally, the increasing gap between processor and off-chip memory speeds has constrained performance of memory-intensive applications. The Single-Chip Message Passing (SCMP) parallel computer sits at the confluence of these trends. SCMP is a tiled architecture consisting of numerous thread-parallel processor and memory nodes connected through a structured interconnection network. Using an interconnection network removes global, ad-hoc wiring that limits scalability and introduces design complexity. However, routing data through general purpose interconnection networks can come at the cost of dedicated bandwidth, longer latency, increased area, and higher power consumption. Understanding the impact architectural decisions have on cost and performance will aid in the eventual adoption of general purpose interconnects. This thesis covers the design and analysis of the on-chip network and its integration with the SCMP system. The result of these efforts is a framework for analyzing on-chip interconnection networks that considers network performance, circuit area, and power consumption. Master of Science 2014-03-14T20:41:52Z 2014-03-14T20:41:52Z 2003-07-23 2003-07-24 2003-08-06 2003-08-06 Thesis etd-07242003-134147 http://hdl.handle.net/10919/34137 http://scholar.lib.vt.edu/theses/available/etd-07242003-134147/ btgthesis.pdf In Copyright http://rightsstatements.org/vocab/InC/1.0/ application/pdf Virginia Tech
collection NDLTD
format Others
sources NDLTD
topic area
virtual channels
SCMP
power
network
router
crossbar switch
single chip computer
message passing
system on chip
spellingShingle area
virtual channels
SCMP
power
network
router
crossbar switch
single chip computer
message passing
system on chip
Gold, Brian
Balancing Performance, Area, and Power in an On-Chip Network
description Several trends can be observed in modern microprocessor design. Architectures have become increasingly complex while design time continues to dwindle. As feature sizes shrink, wire resistance and delay increase, limiting architects from scaling designs centered around a single thread of execution. Where previous decades have focused on exploiting instruction-level parallelism, emerging applications such as streaming media and on-line transaction processing have shown greater thread-level parallelism. Finally, the increasing gap between processor and off-chip memory speeds has constrained performance of memory-intensive applications. The Single-Chip Message Passing (SCMP) parallel computer sits at the confluence of these trends. SCMP is a tiled architecture consisting of numerous thread-parallel processor and memory nodes connected through a structured interconnection network. Using an interconnection network removes global, ad-hoc wiring that limits scalability and introduces design complexity. However, routing data through general purpose interconnection networks can come at the cost of dedicated bandwidth, longer latency, increased area, and higher power consumption. Understanding the impact architectural decisions have on cost and performance will aid in the eventual adoption of general purpose interconnects. This thesis covers the design and analysis of the on-chip network and its integration with the SCMP system. The result of these efforts is a framework for analyzing on-chip interconnection networks that considers network performance, circuit area, and power consumption. === Master of Science
author2 Electrical and Computer Engineering
author_facet Electrical and Computer Engineering
Gold, Brian
author Gold, Brian
author_sort Gold, Brian
title Balancing Performance, Area, and Power in an On-Chip Network
title_short Balancing Performance, Area, and Power in an On-Chip Network
title_full Balancing Performance, Area, and Power in an On-Chip Network
title_fullStr Balancing Performance, Area, and Power in an On-Chip Network
title_full_unstemmed Balancing Performance, Area, and Power in an On-Chip Network
title_sort balancing performance, area, and power in an on-chip network
publisher Virginia Tech
publishDate 2014
url http://hdl.handle.net/10919/34137
http://scholar.lib.vt.edu/theses/available/etd-07242003-134147/
work_keys_str_mv AT goldbrian balancingperformanceareaandpowerinanonchipnetwork
_version_ 1719342940897148928