Credit-based router-to-router flow control is one main link-level flow control mechanism proposed for Networks on Chip (NoCs). Based on network calculus, we analyze its performance and optimal buffer size. To model the feedback control behavior due to credits, we introduce a virtual network service element called flow controller. Then we derive its service curve, and further the system service curve. In addition, we give and prove a theorem that determines the optimal buffer size guaranteeing the maximum system service curve. Moreover, assuming the latency-rate server model for routers, we give closed-form formulas to calculate the flit delay bound and optimal buffer size. Our experiments with real on-chip traffic traces validate that our analysis is correct; delay bounds are tight and the optimal buffer size is exact.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yue QIAN, Zhonghai LU, Wenhua DOU, Qiang DOU, "Analyzing Credit-Based Router-to-Router Flow Control for On-Chip Networks" in IEICE TRANSACTIONS on Electronics,
vol. E92-C, no. 10, pp. 1276-1283, October 2009, doi: 10.1587/transele.E92.C.1276.
Abstract: Credit-based router-to-router flow control is one main link-level flow control mechanism proposed for Networks on Chip (NoCs). Based on network calculus, we analyze its performance and optimal buffer size. To model the feedback control behavior due to credits, we introduce a virtual network service element called flow controller. Then we derive its service curve, and further the system service curve. In addition, we give and prove a theorem that determines the optimal buffer size guaranteeing the maximum system service curve. Moreover, assuming the latency-rate server model for routers, we give closed-form formulas to calculate the flit delay bound and optimal buffer size. Our experiments with real on-chip traffic traces validate that our analysis is correct; delay bounds are tight and the optimal buffer size is exact.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.E92.C.1276/_p
Copy
@ARTICLE{e92-c_10_1276,
author={Yue QIAN, Zhonghai LU, Wenhua DOU, Qiang DOU, },
journal={IEICE TRANSACTIONS on Electronics},
title={Analyzing Credit-Based Router-to-Router Flow Control for On-Chip Networks},
year={2009},
volume={E92-C},
number={10},
pages={1276-1283},
abstract={Credit-based router-to-router flow control is one main link-level flow control mechanism proposed for Networks on Chip (NoCs). Based on network calculus, we analyze its performance and optimal buffer size. To model the feedback control behavior due to credits, we introduce a virtual network service element called flow controller. Then we derive its service curve, and further the system service curve. In addition, we give and prove a theorem that determines the optimal buffer size guaranteeing the maximum system service curve. Moreover, assuming the latency-rate server model for routers, we give closed-form formulas to calculate the flit delay bound and optimal buffer size. Our experiments with real on-chip traffic traces validate that our analysis is correct; delay bounds are tight and the optimal buffer size is exact.},
keywords={},
doi={10.1587/transele.E92.C.1276},
ISSN={1745-1353},
month={October},}
Copy
TY - JOUR
TI - Analyzing Credit-Based Router-to-Router Flow Control for On-Chip Networks
T2 - IEICE TRANSACTIONS on Electronics
SP - 1276
EP - 1283
AU - Yue QIAN
AU - Zhonghai LU
AU - Wenhua DOU
AU - Qiang DOU
PY - 2009
DO - 10.1587/transele.E92.C.1276
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E92-C
IS - 10
JA - IEICE TRANSACTIONS on Electronics
Y1 - October 2009
AB - Credit-based router-to-router flow control is one main link-level flow control mechanism proposed for Networks on Chip (NoCs). Based on network calculus, we analyze its performance and optimal buffer size. To model the feedback control behavior due to credits, we introduce a virtual network service element called flow controller. Then we derive its service curve, and further the system service curve. In addition, we give and prove a theorem that determines the optimal buffer size guaranteeing the maximum system service curve. Moreover, assuming the latency-rate server model for routers, we give closed-form formulas to calculate the flit delay bound and optimal buffer size. Our experiments with real on-chip traffic traces validate that our analysis is correct; delay bounds are tight and the optimal buffer size is exact.
ER -