Optics in Data Center: Improving Scalability and Energy Efficiency

被引:0
|
作者
Cerutti, I. [1 ]
Andriolli, N. [1 ]
Raponi, P. G. [1 ]
Castoldi, P. [1 ]
Liboiron-Ladouceur, O. [2 ]
机构
[1] Scuola Super Sant Anna, Pisa, Italy
[2] McGill Univ, Montreal, PQ, Canada
关键词
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Interconnection networks within modern data centers suffer from bandwidth and scalability limitations, and important power consumption. Optical switching solutions can help to overcome such issues when properly designed and optimized. This paper discusses three areas of improvement beyond state of the art: i) the use of more energy-efficient optical devices to realize high-capacity and energy-efficient optical space switches; ii) the enhancement of their scalability by combining space switching with time switching into a space-time interconnection network architecture (STIA); iii) the planning of an energy-efficient STIA-based network meeting data center bisection bandwidth requirements. Performance assessment shows that i) by combining different switching devices in the same space switch, up to 90% of the power can be saved with respect to conventional SOA-based space switches; ii) STIA allows scalability to increase by a factor of 8, thanks to the combined use of space and time switching domains, while limiting the energy per bit; iii) STIA allows the realization of data center topologies, such as folded Clos and flattened butterfly, whose optimized design can lead to power saving up to 50% with respect to an energy unaware design.
引用
下载
收藏
页数:5
相关论文
共 50 条
  • [31] Scalability and Bandwidth Optimization for Data Center Networks
    Zhang, Teng
    Song, Jiaxing
    Liu, Weidong
    2012 IEEE 14TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS & 2012 IEEE 9TH INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE AND SYSTEMS (HPCC-ICESS), 2012, : 935 - 940
  • [32] The Convergence of Efficiency and Scalability in Ocean Data Platforms
    Manderson, Travis
    Weaver, Matt
    Sinsky, Margaret
    Cook, Mackenzie
    Homsy, Geo
    Nolasco, Emery
    Legge, Ryan
    Srivatsan, Akshay
    Eustace, Alan
    OCEANS 2021: SAN DIEGO - PORTO, 2021,
  • [33] A Testbed and Data Yields for Studying Data Center Energy Efficiency and Reliability
    Van Le, Duc
    Liu, Yingbo
    Wang, Rongrong
    Tan, Rui
    Ngoh, Lek Heng
    PROCEEDINGS OF THE FIRST WORKSHOP ON DATA ACQUISITION TO ANALYSIS (DATA '18), 2018, : 17 - 18
  • [34] Scalability and efficiency in data mining and machine learning
    Miera, Wagner, Jr.
    2019 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2019, : 932 - 932
  • [35] Improving the Efficiency of Server-centric Data Center Network Architectures
    Wang, Ting
    Xia, Yu
    Lin, Dong
    Hamdi, Mounir
    2014 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2014, : 3088 - 3093
  • [36] Improving Energy Efficiency of Buildings Using Data Mining Technologies
    Zucker, Gerhard
    Malinao, Jasmine
    Habib, Usman
    Leber, Thomas
    Preisler, Anita
    Judex, Florian
    2014 IEEE 23RD INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS (ISIE), 2014, : 2664 - 2669
  • [37] Improving energy efficiency in colocation data centers for demand response
    Chi, Ce
    Zhang, Fa
    Ji, Kaixuan
    Marahatta, Avinab
    Liu, Zhiyong
    SUSTAINABLE COMPUTING-INFORMATICS & SYSTEMS, 2021, 29
  • [38] Improving Energy Efficiency of Centrally Controlled Wireless Data Networks
    John A. Stine
    Gustavo de Veciana
    Wireless Networks, 2002, 8 : 681 - 700
  • [39] Improving energy efficiency of centrally controlled wireless data networks
    Stine, JA
    De Veciana, G
    WIRELESS NETWORKS, 2002, 8 (06) : 681 - 700
  • [40] A Survey of FPGA Optimization Methods for Data Center Energy Efficiency
    Tibaldi, Mattia
    Pilato, Christian
    IEEE TRANSACTIONS ON SUSTAINABLE COMPUTING, 2023, 8 (03): : 343 - 362