TY - JOUR
T1 - Ephedrine QoS
T2 - An antidote to slow, congested, bufferless NoCs
AU - Fang, Juan
AU - Yao, Zhicheng
AU - Sui, Xiufeng
AU - Bao, Yungang
N1 - Publisher Copyright:
© 2014 Juan Fang et al.
PY - 2014
Y1 - 2014
N2 - 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%.
AB - 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%.
UR - http://www.scopus.com/inward/record.url?scp=84933059867&partnerID=8YFLogxK
U2 - 10.1155/2014/691865
DO - 10.1155/2014/691865
M3 - Article
C2 - 25250386
AN - SCOPUS:84933059867
SN - 2356-6140
VL - 2014
JO - The Scientific World Journal
JF - The Scientific World Journal
M1 - 691865
ER -