Cargando…
Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs
Datacenters consolidate diverse applications to improve utilization. However when multiple applications are colocated on such platforms, contention for shared resources like networks-on-chip (NoCs) can degrade the performance of latency-critical online services (high-priority applications). Recently...
Autores principales: | , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Hindawi Publishing Corporation
2014
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4163332/ https://www.ncbi.nlm.nih.gov/pubmed/25250386 http://dx.doi.org/10.1155/2014/691865 |
_version_ | 1782334790013288448 |
---|---|
author | Fang, Juan Yao, Zhicheng Sui, Xiufeng Bao, Yungang |
author_facet | Fang, Juan Yao, Zhicheng Sui, Xiufeng Bao, Yungang |
author_sort | Fang, Juan |
collection | PubMed |
description | Datacenters consolidate diverse applications to improve utilization. However when multiple applications are colocated on such platforms, contention for shared resources like networks-on-chip (NoCs) can degrade the performance of latency-critical online services (high-priority applications). Recently proposed bufferless NoCs (Nychis et al.) have the advantages of requiring less area and power, but they pose challenges in quality-of-service (QoS) support, which usually relies on buffer-based virtual channels (VCs). We propose QBLESS, a QoS-aware bufferless NoC scheme for datacenters. QBLESS consists of two components: a routing mechanism (QBLESS-R) that can substantially reduce flit deflection for high-priority applications and a congestion-control mechanism (QBLESS-CC) that guarantees performance for high-priority applications and improves overall system throughput. We use trace-driven simulation to model a 64-core system, finding that, when compared to BLESS, a previous state-of-the-art bufferless NoC design, QBLESS, improves performance of high-priority applications by an average of 33.2% and reduces network-hops by an average of 42.8%. |
format | Online Article Text |
id | pubmed-4163332 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2014 |
publisher | Hindawi Publishing Corporation |
record_format | MEDLINE/PubMed |
spelling | pubmed-41633322014-09-23 Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs Fang, Juan Yao, Zhicheng Sui, Xiufeng Bao, Yungang ScientificWorldJournal Research Article Datacenters consolidate diverse applications to improve utilization. However when multiple applications are colocated on such platforms, contention for shared resources like networks-on-chip (NoCs) can degrade the performance of latency-critical online services (high-priority applications). Recently proposed bufferless NoCs (Nychis et al.) have the advantages of requiring less area and power, but they pose challenges in quality-of-service (QoS) support, which usually relies on buffer-based virtual channels (VCs). We propose QBLESS, a QoS-aware bufferless NoC scheme for datacenters. QBLESS consists of two components: a routing mechanism (QBLESS-R) that can substantially reduce flit deflection for high-priority applications and a congestion-control mechanism (QBLESS-CC) that guarantees performance for high-priority applications and improves overall system throughput. We use trace-driven simulation to model a 64-core system, finding that, when compared to BLESS, a previous state-of-the-art bufferless NoC design, QBLESS, improves performance of high-priority applications by an average of 33.2% and reduces network-hops by an average of 42.8%. Hindawi Publishing Corporation 2014 2014-08-28 /pmc/articles/PMC4163332/ /pubmed/25250386 http://dx.doi.org/10.1155/2014/691865 Text en Copyright © 2014 Juan Fang et al. https://creativecommons.org/licenses/by/3.0/ This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. |
spellingShingle | Research Article Fang, Juan Yao, Zhicheng Sui, Xiufeng Bao, Yungang Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title | Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title_full | Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title_fullStr | Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title_full_unstemmed | Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title_short | Ephedrine QoS: An Antidote to Slow, Congested, Bufferless NoCs |
title_sort | ephedrine qos: an antidote to slow, congested, bufferless nocs |
topic | Research Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4163332/ https://www.ncbi.nlm.nih.gov/pubmed/25250386 http://dx.doi.org/10.1155/2014/691865 |
work_keys_str_mv | AT fangjuan ephedrineqosanantidotetoslowcongestedbufferlessnocs AT yaozhicheng ephedrineqosanantidotetoslowcongestedbufferlessnocs AT suixiufeng ephedrineqosanantidotetoslowcongestedbufferlessnocs AT baoyungang ephedrineqosanantidotetoslowcongestedbufferlessnocs |