In-kernel integration of operating system and infiniband functions for high performance computing clusters: A DSM example

被引:4
|
作者
Liss, L [1 ]
Birk, Y
Schuster, A
机构
[1] Technion Israel Inst Technol, Dept Elect Engn, IL-32000 Haifa, Israel
[2] Technion Israel Inst Technol, Dept Comp Sci, IL-32000 Haifa, Israel
关键词
hardware/software interfaces; high-speed networks; distributed shared memory; parallel computing;
D O I
10.1109/TPDS.2005.111
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
The Infiniband (IB) System Area Network (SAN) enables applications to access hardware directly from user level, reducing the overhead of user-kernel crossings during data transfer. However, distributed applications that exhibit close coupling between network and OS services may benefit from accessing IB from the kernel through IB's native Verbs interface, which permits tight integration of these services. We assess this approach using a sequential-consistency Distributed Shared Memory (DSM) system as an example. We first develop primitives that abstract the low-level communication and kernel details, and efficiently serve the application's communication, memory, and scheduling needs. Next, we combine the primitives to form a kernel DSM protocol. The approach is evaluated using our full-fledged Linux kernel DSM implementation over Infiniband. We show that overheads are reduced substantially, and overall application performance is improved in terms of both absolute execution time and scalability relative to an entirely user level implementation.
引用
收藏
页码:830 / 840
页数:11
相关论文
共 13 条
  • [1] The MOSIX multicomputer operating system for high performance cluster computing
    Barak, A
    La'adan, O
    FUTURE GENERATION COMPUTER SYSTEMS, 1998, 13 (4-5) : 361 - 372
  • [2] Kerrighed:: A single system image cluster operating system for high performance computing
    Morin, C
    Lottiaux, R
    Vallée, G
    Gallard, P
    Utard, G
    Badrinath, R
    Rilling, L
    EURO-PAR 2003 PARALLEL PROCESSING, PROCEEDINGS, 2003, 2790 : 1291 - 1294
  • [3] Advanced System Integration for High Performance Computing with Liquid Cooling
    Hung, Jeng-Nan
    Li, Hung-Chi
    Lin, Po-Fan
    Ku, Terry
    Yu, C.H.
    Yee, K.C.
    Yu, Doug C.H.
    Proceedings - Electronic Components and Technology Conference, 2021, 2021-June : 105 - 111
  • [4] Advanced System Integration for High Performance Computing with Liquid Cooling
    Hung, Jeng-Nan
    Li, Hung-Chi
    Lin, Po-Fan
    Ku, Terry
    Yu, C. H.
    Yee, Kc
    Yu, Doug C. H.
    IEEE 71ST ELECTRONIC COMPONENTS AND TECHNOLOGY CONFERENCE (ECTC 2021), 2021, : 105 - 111
  • [5] A high performance computing system for medical imaging in the remote operating room
    Kawasaki, Y
    Ino, F
    Mizutani, Y
    Fujimoto, N
    Sasama, T
    Sato, Y
    Tamura, S
    Hagihara, K
    HIGH PERFORMANCE COMPUTING - HIPC 2003, 2003, 2913 : 162 - 173
  • [6] Using a single address space operating system for distributed computing and high performance
    Skousen, A
    Miller, D
    1999 IEEE INTERNATIONAL PERFORMANCE, COMPUTING AND COMMUNICATIONS CONFERENCE, 1999, : 8 - 14
  • [7] The cluster file system: Integration of high performance communication and I/O in clusters
    Cristaldi, R
    Iannello, G
    Delfino, F
    CCGRID 2002: 2ND IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER COMPUTING AND THE GRID, PROCEEDINGS, 2002, : 173 - 182
  • [8] Making Profit with ALBATROSS: A Runtime System for Heterogeneous High-Performance-Computing Clusters
    Hoenig, Timo
    Eibel, Christopher
    Wagenhaeuser, Adam
    Wagner, Maximilian
    Schroeder-Preikschat, Wolfgang
    HPDC '18: PROCEEDINGS OF THE 27TH INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE PARALLEL AND DISTRIBUTED COMPUTING: POSTERS/DOCTORAL CONSORTIUM, 2018, : 11 - 12
  • [9] A Virtualization Based Elastic Model for High Performance Computing Clusters in a Networked Control System
    Xu Lijun
    Fei Minrui
    Yu Wei
    Song Yang
    Du Dajun
    2011 30TH CHINESE CONTROL CONFERENCE (CCC), 2011, : 6527 - 6532
  • [10] Multi-wavelength transceiver integration on SOI for high-performance computing system applications
    Aalto, Timo
    Harjanne, Mikko
    Ylinen, Sami
    Kapulainen, Markku
    Vehmas, Tapani
    Cherchi, Matteo
    Neumeyr, Christian
    Ortsiefer, Markus
    Malacarne, Antonio
    OPTICAL INTERCONNECTS XV, 2015, 9368