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
Um data center convencional que consiste em servidores monolíticos é confrontado com limitações, incluindo falta de flexibilidade operacional, baixa utilização de recursos, baixa capacidade de manutenção, etc. A desagregação de recursos é uma solução promissora para resolver os problemas acima. Propomos um conceito de arquitetura desagregada de data center em nuvem chamada Flow-in-Cloud (FiC), que permite que um sistema de cluster de computador existente expanda um pool de aceleradores por meio de uma rede de alta velocidade. FlowOS-RM gerencia todos os recursos do pool e implanta um trabalho do usuário em uma fatia construída dinamicamente de acordo com uma solicitação do usuário. Esta fatia consiste em nós de computação e aceleradores, onde cada acelerador é anexado ao nó de computação correspondente. Este artigo demonstra a viabilidade do FiC em um experimento de prova de conceito executando um aplicativo de aprendizado profundo distribuído no sistema protótipo. O resultado garante com sucesso a aplicabilidade do sistema proposto.
Ryousei TAKANO
National Institute of Advanced Industrial Science and Technology (AIST)
Kuniyasu SUZAKI
National Institute of Advanced Industrial Science and Technology (AIST)
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
Ryousei TAKANO, Kuniyasu SUZAKI, "Disaggregated Accelerator Management System for Cloud Data Centers" in IEICE TRANSACTIONS on Information,
vol. E104-D, no. 3, pp. 465-468, March 2021, doi: 10.1587/transinf.2020EDL8040.
Abstract: A conventional data center that consists of monolithic-servers is confronted with limitations including lack of operational flexibility, low resource utilization, low maintainability, etc. Resource disaggregation is a promising solution to address the above issues. We propose a concept of disaggregated cloud data center architecture called Flow-in-Cloud (FiC) that enables an existing cluster computer system to expand an accelerator pool through a high-speed network. FlowOS-RM manages the entire pool resources, and deploys a user job on a dynamically constructed slice according to a user request. This slice consists of compute nodes and accelerators where each accelerator is attached to the corresponding compute node. This paper demonstrates the feasibility of FiC in a proof of concept experiment running a distributed deep learning application on the prototype system. The result successfully warrants the applicability of the proposed system.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDL8040/_p
Copiar
@ARTICLE{e104-d_3_465,
author={Ryousei TAKANO, Kuniyasu SUZAKI, },
journal={IEICE TRANSACTIONS on Information},
title={Disaggregated Accelerator Management System for Cloud Data Centers},
year={2021},
volume={E104-D},
number={3},
pages={465-468},
abstract={A conventional data center that consists of monolithic-servers is confronted with limitations including lack of operational flexibility, low resource utilization, low maintainability, etc. Resource disaggregation is a promising solution to address the above issues. We propose a concept of disaggregated cloud data center architecture called Flow-in-Cloud (FiC) that enables an existing cluster computer system to expand an accelerator pool through a high-speed network. FlowOS-RM manages the entire pool resources, and deploys a user job on a dynamically constructed slice according to a user request. This slice consists of compute nodes and accelerators where each accelerator is attached to the corresponding compute node. This paper demonstrates the feasibility of FiC in a proof of concept experiment running a distributed deep learning application on the prototype system. The result successfully warrants the applicability of the proposed system.},
keywords={},
doi={10.1587/transinf.2020EDL8040},
ISSN={1745-1361},
month={March},}
Copiar
TY - JOUR
TI - Disaggregated Accelerator Management System for Cloud Data Centers
T2 - IEICE TRANSACTIONS on Information
SP - 465
EP - 468
AU - Ryousei TAKANO
AU - Kuniyasu SUZAKI
PY - 2021
DO - 10.1587/transinf.2020EDL8040
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E104-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2021
AB - A conventional data center that consists of monolithic-servers is confronted with limitations including lack of operational flexibility, low resource utilization, low maintainability, etc. Resource disaggregation is a promising solution to address the above issues. We propose a concept of disaggregated cloud data center architecture called Flow-in-Cloud (FiC) that enables an existing cluster computer system to expand an accelerator pool through a high-speed network. FlowOS-RM manages the entire pool resources, and deploys a user job on a dynamically constructed slice according to a user request. This slice consists of compute nodes and accelerators where each accelerator is attached to the corresponding compute node. This paper demonstrates the feasibility of FiC in a proof of concept experiment running a distributed deep learning application on the prototype system. The result successfully warrants the applicability of the proposed system.
ER -