End-to-end network latency has become an important issue for parallel application on large-scale high performance computing (HPC) systems. It has been reported that randomly-connected inter-switch networks can lower the end-to-end network latency. This latency reduction is established in exchange for a large amount of routing information. That is, minimal routing on irregular networks is achieved by using routing tables for all destinations in the networks. In this work, a novel distributed routing method called LOREN (Layout-Oriented Routing with Entries for Neighbors) to achieve low-latency with a small routing table is proposed for irregular networks whose link length is limited. The routing tables contain both physically and topologically nearby neighbor nodes to ensure livelock-freedom and a small number of hops between nodes. Experimental results show that LOREN reduces the average latencies by 5.8% and improves the network throughput by up to 62% compared with a conventional compact routing method. Moreover, the number of required routing table entries is reduced by up to 91%, which improves scalability and flexibility for implementation.
Ryuta KAWANO
Keio University
Hiroshi NAKAHARA
Keio University
Ikki FUJIWARA
National Institute of Information and Communications Technology
Hiroki MATSUTANI
Keio University
Michihiro KOIBUCHI
National Institute of Informatics
Hideharu AMANO
Keio University
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
Ryuta KAWANO, Hiroshi NAKAHARA, Ikki FUJIWARA, Hiroki MATSUTANI, Michihiro KOIBUCHI, Hideharu AMANO, "A Layout-Oriented Routing Method for Low-Latency HPC Networks" in IEICE TRANSACTIONS on Information,
vol. E100-D, no. 12, pp. 2796-2807, December 2017, doi: 10.1587/transinf.2017PAP0019.
Abstract: End-to-end network latency has become an important issue for parallel application on large-scale high performance computing (HPC) systems. It has been reported that randomly-connected inter-switch networks can lower the end-to-end network latency. This latency reduction is established in exchange for a large amount of routing information. That is, minimal routing on irregular networks is achieved by using routing tables for all destinations in the networks. In this work, a novel distributed routing method called LOREN (Layout-Oriented Routing with Entries for Neighbors) to achieve low-latency with a small routing table is proposed for irregular networks whose link length is limited. The routing tables contain both physically and topologically nearby neighbor nodes to ensure livelock-freedom and a small number of hops between nodes. Experimental results show that LOREN reduces the average latencies by 5.8% and improves the network throughput by up to 62% compared with a conventional compact routing method. Moreover, the number of required routing table entries is reduced by up to 91%, which improves scalability and flexibility for implementation.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017PAP0019/_p
Copy
@ARTICLE{e100-d_12_2796,
author={Ryuta KAWANO, Hiroshi NAKAHARA, Ikki FUJIWARA, Hiroki MATSUTANI, Michihiro KOIBUCHI, Hideharu AMANO, },
journal={IEICE TRANSACTIONS on Information},
title={A Layout-Oriented Routing Method for Low-Latency HPC Networks},
year={2017},
volume={E100-D},
number={12},
pages={2796-2807},
abstract={End-to-end network latency has become an important issue for parallel application on large-scale high performance computing (HPC) systems. It has been reported that randomly-connected inter-switch networks can lower the end-to-end network latency. This latency reduction is established in exchange for a large amount of routing information. That is, minimal routing on irregular networks is achieved by using routing tables for all destinations in the networks. In this work, a novel distributed routing method called LOREN (Layout-Oriented Routing with Entries for Neighbors) to achieve low-latency with a small routing table is proposed for irregular networks whose link length is limited. The routing tables contain both physically and topologically nearby neighbor nodes to ensure livelock-freedom and a small number of hops between nodes. Experimental results show that LOREN reduces the average latencies by 5.8% and improves the network throughput by up to 62% compared with a conventional compact routing method. Moreover, the number of required routing table entries is reduced by up to 91%, which improves scalability and flexibility for implementation.},
keywords={},
doi={10.1587/transinf.2017PAP0019},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - A Layout-Oriented Routing Method for Low-Latency HPC Networks
T2 - IEICE TRANSACTIONS on Information
SP - 2796
EP - 2807
AU - Ryuta KAWANO
AU - Hiroshi NAKAHARA
AU - Ikki FUJIWARA
AU - Hiroki MATSUTANI
AU - Michihiro KOIBUCHI
AU - Hideharu AMANO
PY - 2017
DO - 10.1587/transinf.2017PAP0019
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2017
AB - End-to-end network latency has become an important issue for parallel application on large-scale high performance computing (HPC) systems. It has been reported that randomly-connected inter-switch networks can lower the end-to-end network latency. This latency reduction is established in exchange for a large amount of routing information. That is, minimal routing on irregular networks is achieved by using routing tables for all destinations in the networks. In this work, a novel distributed routing method called LOREN (Layout-Oriented Routing with Entries for Neighbors) to achieve low-latency with a small routing table is proposed for irregular networks whose link length is limited. The routing tables contain both physically and topologically nearby neighbor nodes to ensure livelock-freedom and a small number of hops between nodes. Experimental results show that LOREN reduces the average latencies by 5.8% and improves the network throughput by up to 62% compared with a conventional compact routing method. Moreover, the number of required routing table entries is reduced by up to 91%, which improves scalability and flexibility for implementation.
ER -