The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
O sistema multi-FPGA conhecido como sistema Flow-in-Cloud (FiC), é composto por FPGAs de médio porte que são diretamente interconectados por links seriais de alta velocidade. Atualmente, o FiC está sendo desenvolvido como um servidor para computação de borda multiacesso (MEC), que é uma das principais tecnologias do 5G. Como as aplicações do MEC às vezes são críticas em termos de tempo, uma rede estática de multiplexação por divisão de tempo (STDM) tem sido usada no FiC. Contudo, a rede STDM apresenta a desvantagem de diminuir a utilização do link, especialmente sob tráfego leve. Para resolver este problema, propomos um roteador híbrido que combina comutação de pacotes para comunicação de baixa prioridade e STDM para comunicação de alta prioridade. Em nossa rede híbrida, a comutação de pacotes utiliza slots que não são utilizados pelo STDM; portanto, a comunicação de melhor esforço por comutação de pacotes e a comunicação de garantia de QoS pelo STDM podem ser usadas simultaneamente. Além disso, para melhorar a utilização de cada link sob uma carga de tráfego de rede baixa, propomos um algoritmo de comutação dinâmica de comunicação. Em nosso algoritmo, cada roteador monitora as métricas de carga da rede e, de acordo com as métricas, as tarefas críticas de tempo selecionam o STDM de acordo com as métricas apenas quando ocorre congestionamento. Isso pode garantir a garantia de QoS e a utilização eficiente de cada link com uma pequena sobrecarga de recursos. Em nossa avaliação, o algoritmo dinâmico foi até 24.6% mais rápido no tempo de execução com alta carga de rede em comparação com a comutação de pacotes em um sistema multi-FPGA real com 24 placas.
Tomoki SHIMIZU
Keio University
Kohei ITO
Keio University
Kensuke IIZUKA
Keio University
Kazuei HIRONAKA
Keio University
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.
Copiar
Tomoki SHIMIZU, Kohei ITO, Kensuke IIZUKA, Kazuei HIRONAKA, Hideharu AMANO, "The Implementation of a Hybrid Router and Dynamic Switching Algorithm on a Multi-FPGA System" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 12, pp. 2008-2018, December 2022, doi: 10.1587/transinf.2022PAP0009.
Abstract: The multi-FPGA system known as, the Flow-in-Cloud (FiC) system, is composed of mid-range FPGAs that are directly interconnected by high-speed serial links. FiC is currently being developed as a server for multi-access edge computing (MEC), which is one of the core technologies of 5G. Because the applications of MEC are sometimes timing-critical, a static time division multiplexing (STDM) network has been used on FiC. However, the STDM network exhibits the disadvantage of decreasing link utilization, especially under light traffic. To solve this problem, we propose a hybrid router that combines packet switching for low-priority communication and STDM for high-priority communication. In our hybrid network, the packet switching uses slots that are unused by the STDM; therefore, best-effort communication by packet switching and QoS guarantee communication by the STDM can be used simultaneously. Furthermore, to improve each link utilization under a low network traffic load, we propose a dynamic communication switching algorithm. In our algorithm, each router monitors the network load metrics, and according to the metrics, timing-critical tasks select the STDM according to the metrics only when congestion occurs. This can achieve both QoS guarantee and efficient utilization of each link with a small resource overhead. In our evaluation, the dynamic algorithm was up to 24.6% faster on the execution time with a high network load compared to the packet switching on a real multi-FPGA system with 24 boards.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022PAP0009/_p
Copiar
@ARTICLE{e105-d_12_2008,
author={Tomoki SHIMIZU, Kohei ITO, Kensuke IIZUKA, Kazuei HIRONAKA, Hideharu AMANO, },
journal={IEICE TRANSACTIONS on Information},
title={The Implementation of a Hybrid Router and Dynamic Switching Algorithm on a Multi-FPGA System},
year={2022},
volume={E105-D},
number={12},
pages={2008-2018},
abstract={The multi-FPGA system known as, the Flow-in-Cloud (FiC) system, is composed of mid-range FPGAs that are directly interconnected by high-speed serial links. FiC is currently being developed as a server for multi-access edge computing (MEC), which is one of the core technologies of 5G. Because the applications of MEC are sometimes timing-critical, a static time division multiplexing (STDM) network has been used on FiC. However, the STDM network exhibits the disadvantage of decreasing link utilization, especially under light traffic. To solve this problem, we propose a hybrid router that combines packet switching for low-priority communication and STDM for high-priority communication. In our hybrid network, the packet switching uses slots that are unused by the STDM; therefore, best-effort communication by packet switching and QoS guarantee communication by the STDM can be used simultaneously. Furthermore, to improve each link utilization under a low network traffic load, we propose a dynamic communication switching algorithm. In our algorithm, each router monitors the network load metrics, and according to the metrics, timing-critical tasks select the STDM according to the metrics only when congestion occurs. This can achieve both QoS guarantee and efficient utilization of each link with a small resource overhead. In our evaluation, the dynamic algorithm was up to 24.6% faster on the execution time with a high network load compared to the packet switching on a real multi-FPGA system with 24 boards.},
keywords={},
doi={10.1587/transinf.2022PAP0009},
ISSN={1745-1361},
month={December},}
Copiar
TY - JOUR
TI - The Implementation of a Hybrid Router and Dynamic Switching Algorithm on a Multi-FPGA System
T2 - IEICE TRANSACTIONS on Information
SP - 2008
EP - 2018
AU - Tomoki SHIMIZU
AU - Kohei ITO
AU - Kensuke IIZUKA
AU - Kazuei HIRONAKA
AU - Hideharu AMANO
PY - 2022
DO - 10.1587/transinf.2022PAP0009
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2022
AB - The multi-FPGA system known as, the Flow-in-Cloud (FiC) system, is composed of mid-range FPGAs that are directly interconnected by high-speed serial links. FiC is currently being developed as a server for multi-access edge computing (MEC), which is one of the core technologies of 5G. Because the applications of MEC are sometimes timing-critical, a static time division multiplexing (STDM) network has been used on FiC. However, the STDM network exhibits the disadvantage of decreasing link utilization, especially under light traffic. To solve this problem, we propose a hybrid router that combines packet switching for low-priority communication and STDM for high-priority communication. In our hybrid network, the packet switching uses slots that are unused by the STDM; therefore, best-effort communication by packet switching and QoS guarantee communication by the STDM can be used simultaneously. Furthermore, to improve each link utilization under a low network traffic load, we propose a dynamic communication switching algorithm. In our algorithm, each router monitors the network load metrics, and according to the metrics, timing-critical tasks select the STDM according to the metrics only when congestion occurs. This can achieve both QoS guarantee and efficient utilization of each link with a small resource overhead. In our evaluation, the dynamic algorithm was up to 24.6% faster on the execution time with a high network load compared to the packet switching on a real multi-FPGA system with 24 boards.
ER -