Entry Chandra:1994:SPM from sigplan1990.bib

Last update: Thu Apr 12 03:37:15 MDT 2012                Valid HTML 3.2!

Index sections

Top | Symbols | Numbers | Math | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

BibTeX entry

@Article{Chandra:1994:SPM,
  author =       "Rohit Chandra and Scott Devine and Ben Verghese and
                 Anoop Gupta and Mendel Rosenblum",
  title =        "Scheduling and page migration for multiprocessor
                 compute servers",
  journal =      j-SIGPLAN,
  volume =       "29",
  number =       "11",
  pages =        "12--24",
  month =        nov,
  year =         "1994",
  CODEN =        "SINODQ",
  ISSN =         "0362-1340 (print), 1523-2867 (print), 1558-1160 (electronic)",
  ISSN-L =       "0362-1340",
  bibdate =      "Sun Dec 14 09:16:57 MST 2003",
  bibsource =    "http://portal.acm.org/; http://www.acm.org/pubs/toc/",
  URL =          "http://www.acm.org:80/pubs/citations/proceedings/asplos/195473/p12-chandra/",
  abstract =     "Several cache-coherent shared-memory multiprocessors
                 have been developed that are scalable and offer a very
                 tight coupling between the processing resources. They
                 are therefore quite attractive for use as compute
                 servers for multiprogramming and parallel application
                 workloads. Process scheduling and memory management,
                 however, remain challenging due to the distributed main
                 memory found on such machines. This paper examines the
                 effects of OS scheduling and page migration policies on
                 the performance of such compute servers. Our
                 experiments are done on the Stanford DASH, a
                 distributed-memory cache-coherent multiprocessor. We
                 show that for our multiprogramming workloads consisting
                 of sequential jobs, the traditional Unix scheduling
                 policy does very poorly. In contrast, a policy
                 incorporating cluster and cache affinity along with a
                 simple page-migration algorithm offers up to two-fold
                 performance improvement. For our workloads consisting
                 of multiple parallel applications, we compare
                 space-sharing policies that divide the processors among
                 the applications to time-slicing policies such as
                 standard Unix or gang scheduling. We show that
                 space-sharing policies can achieve better processor
                 utilization due to the operating point effect, but
                 time-slicing policies benefit strongly from user-level
                 data distribution. Our initial experience with
                 automatic page migration suggests that policies based
                 only on TLB miss information can be quite effective,
                 and useful for addressing the data distribution
                 problems of space-sharing schedulers.",
  acknowledgement = ack-nhfb,
  classification = "C5440 (Multiprocessing systems); C6120 (File
                 organisation); C6150J (Operating systems); C6150N
                 (Distributed systems software)",
  conflocation = "San Jose, CA, USA; 4-7 Oct. 1994",
  conftitle =    "Sixth International Conference on Architectural
                 Support for Programming Languages and Operating Systems
                 (ASPLOS-VI)",
  corpsource =   "Comput. Syst. Lab., Stanford Univ., CA, USA",
  keywords =     "algorithms; cache affinity; cache coherent shared
                 memory multiprocessors; design; distributed main
                 memory; distributed memory cache coherent
                 multiprocessor; distributed memory systems;
                 experimentation; gang scheduling; measurement; memory
                 management; multiple parallel applications;
                 multiprocessor compute servers; multiprogramming;
                 operating point effect; OS scheduling; page migration;
                 paged storage; parallel application workloads;
                 performance; performance improvement; process
                 scheduling; processing resources; processor scheduling;
                 processor utilization; scheduling policy; sequential
                 jobs; shared memory systems; space sharing policies;
                 Stanford DASH; storage management; theory; time slicing
                 policies; Unix scheduling policy",
  sponsororg =   "ACM; IEEE Comput. Soc",
  subject =      "{\bf D.4.1} Software, OPERATING SYSTEMS, Process
                 Management, Scheduling.",
  treatment =    "P Practical",
}

Related entries