Power management of online data-intensive services

Much of the success of the Internet services model can be attributed to the popularity of a class of workloads that we call Online Data-Intensive (OLDI) services. These workloads perform significant computing over massive data sets per user request but, unlike their offline counterparts (such as Map...

Full description

Saved in:
Bibliographic Details
Published in:2011 38th Annual International Symposium on Computer Architecture (ISCA) pp. 319 - 330
Main Authors: Meisner, David, Sadler, Christopher M., Barroso, Luiz André, Weber, Wolf-Dietrich, Wenisch, Thomas F.
Format: Conference Proceeding
Language:English
Published: New York, NY, USA ACM 04.06.2011
IEEE
Series:ACM Conferences
Subjects:
Abstracts
Computer systems organization > Architectures > Distributed architectures > Client-server architectures
Information systems > Data management systems > Middleware for databases > Application servers
Information systems > Data management systems > Middleware for databases > Database web servers
Servers
power management
servers
ISBN:9781450304726, 1450304729
ISSN:1063-6897
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Much of the success of the Internet services model can be attributed to the popularity of a class of workloads that we call Online Data-Intensive (OLDI) services. These workloads perform significant computing over massive data sets per user request but, unlike their offline counterparts (such as MapReduce computations), they require responsiveness in the sub-second time scale at high request rates. Large search products, online advertising, and machine translation are examples of workloads in this class. Although the load in OLDI services can vary widely during the day, their energy consumption sees little variance due to the lack of energy proportionality of the underlying machinery. The scale and latency sensitivity of OLDI workloads also make them a challenging target for power management techniques. We investigate what, if anything, can be done to make OLDI systems more energy-proportional. Specifically, we evaluate the applicability of active and idle low-power modes to reduce the power consumed by the primary server components (processor, memory, and disk), while maintaining tight response time constraints, particularly on 95th-percentile latency. Using Web search as a representative example of this workload class, we first characterize a production Web search workload at cluster-wide scale. We provide a fine-grain characterization and expose the opportunity for power savings using low-power modes of each primary server component. Second, we develop and validate a performance model to evaluate the impact of processor- and memory-based low-power modes on the search latency distribution and consider the benefit of current and foreseeable low-power modes. Our results highlight the challenges of power management for this class of workloads. In contrast to other server workloads, for which idle low-power modes have shown great promise, for OLDI workloads we find that energy-proportionality with acceptable query latency can only be achieved using coordinated, full-system active low-power modes.
AbstractList Much of the success of the Internet services model can be attributed to the popularity of a class of workloads that we call Online Data-Intensive (OLDI) services. These workloads perform significant computing over massive data sets per user request but, unlike their offline counterparts (such as MapReduce computations), they require responsiveness in the sub-second time scale at high request rates. Large search products, online advertising, and machine translation are examples of workloads in this class. Although the load in OLDI services can vary widely during the day, their energy consumption sees little variance due to the lack of energy proportionality of the underlying machinery. The scale and latency sensitivity of OLDI workloads also make them a challenging target for power management techniques. We investigate what, if anything, can be done to make OLDI systems more energy-proportional. Specifically, we evaluate the applicability of active and idle low-power modes to reduce the power consumed by the primary server components (processor, memory, and disk), while maintaining tight response time constraints, particularly on 95th-percentile latency. Using Web search as a representative example of this workload class, we first characterize a production Web search workload at cluster-wide scale. We provide a fine-grain characterization and expose the opportunity for power savings using low-power modes of each primary server component. Second, we develop and validate a performance model to evaluate the impact of processor- and memory-based low-power modes on the search latency distribution and consider the benefit of current and foreseeable low-power modes. Our results highlight the challenges of power management for this class of workloads. In contrast to other server workloads, for which idle low-power modes have shown great promise, for OLDI workloads we find that energy-proportionality with acceptable query latency can only be achieved using coordinated, full-system active low-power modes.
Much of the success of the Internet services model can be attributed to the popularity of a class of workloads that we call Online Data-Intensive (OLDI) services. These work-loads perform significant computing over massive data sets per user request but, unlike their offline counterparts (such as MapReduce computations), they require responsiveness in the sub-second time scale at high request rates. Large search products, online advertising, and machine translation are examples of workloads in this class. Although the load in OLDI services can vary widely during the day, their energy consumption sees little variance due to the lack of energy proportionality of the underlying machinery. The scale and latency sensitivity of OLDI workloads also make them a challenging target for power management techniques. We investigate what, if anything, can be done to make OLDI systems more energy-proportional. Specifically, we evaluate the applicability of active and idle low-power modes to reduce the power consumed by the primary server components (processor, memory, and disk), while maintaining tight response time constraints, particularly on 95th-percentile latency. Using Web search as a representative example of this workload class, we first characterize a production Web search workload at cluster-wide scale. We provide a fine-grain characterization and expose the opportunity for power savings using low-power modes of each primary server component. Second, we develop and validate a performance model to evaluate the impact of processor- and memory-based low-power modes on the search latency distribution and consider the benefit of current and foreseeable low-power modes. Our results highlight the challenges of power management for this class of workloads. In contrast to other server workloads, for which idle low-power modes have shown great promise, for OLDI workloads we find that energy-proportionality with acceptable query latency can only be achieved using coordinated, full-system active low-power modes.
Author Meisner, David
Sadler, Christopher M.
Barroso, Luiz André
Weber, Wolf-Dietrich
Wenisch, Thomas F.
Author_xml – sequence: 1
  givenname: David
  surname: Meisner
  fullname: Meisner, David
  email: meisner@umich.edu
  organization: The University of Michigan, Ann Arbor, MI, USA
– sequence: 2
  givenname: Christopher M.
  surname: Sadler
  fullname: Sadler, Christopher M.
  email: cmsadler@gmail.com
  organization: Google, Inc., Mountain View, CA, USA
– sequence: 3
  givenname: Luiz André
  surname: Barroso
  fullname: Barroso, Luiz André
  email: luiz@google.com
  organization: Google, Inc., Mountain View, CA, USA
– sequence: 4
  givenname: Wolf-Dietrich
  surname: Weber
  fullname: Weber, Wolf-Dietrich
  email: wolf@google.com
  organization: Google, Inc., Mountain View, CA, USA
– sequence: 5
  givenname: Thomas F.
  surname: Wenisch
  fullname: Wenisch, Thomas F.
  email: twenisch@umich.edu
  organization: The University of Michigan, Ann Arbor, MI, USA
BookMark eNqNkD1PwzAURY0oEqVkZmDJyJLib8cjqiggVYKhu_WcPiND46A4KuLfk6idmHjL1dPRucO9IrPUJSTkhtElY1LdczqelsspGRVnpLCmHgEVVBquz__8MzJnVItK19ZckiLnj8m3RkrF54S_dd_Yly0keMcW01B2oezSPiYsdzBAFdOAKccDlhn7Q2wwX5OLAPuMxSkXZLt-3K6eq83r08vqYVMBl2aoBGqlmAke6lqpHfrGBBVqAdoD2h0oHrwSElSwxmLQjVJcMmqbURTgxYLcHmsjIrqvPrbQ_zgtqDG6HunySKFpne-6z-wYddM-7rSPO-3jfB8xjMLdPwXxC5m3Yac
ContentType Conference Proceeding
Copyright 2011 ACM
Copyright_xml – notice: 2011 ACM
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1145/2000064.2000103
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList

Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Xplore : IEEE Electronic Library (IEL) [unlimited simultaenous users]
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 9781450304726
1450304729
EndPage 330
ExternalDocumentID 6307768
Genre orig-research
GroupedDBID 6IE
6IF
6IG
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ACM
ADFMO
ADPZR
ALMA_UNASSIGNED_HOLDINGS
APO
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
GUFHI
IEGSK
IERZE
IJVOP
LHSKQ
OCL
RIB
RIC
RIE
RIL
RIO
23M
29F
29O
ACGFS
ADZIZ
CHZPO
IPLJI
M43
ZY4
ID FETCH-LOGICAL-a247t-3e65517fba8855debc7f5f83a6bae9da52fb534a5f979ef6c5524109c3e63ab3
IEDL.DBID RIE
ISBN 9781450304726
1450304729
ISICitedReferencesCount 175
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000292709800028&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1063-6897
IngestDate Wed Aug 27 03:28:42 EDT 2025
Wed Jan 31 06:47:39 EST 2024
Wed Jan 08 03:36:12 EST 2025
IsPeerReviewed false
IsScholarly true
Keywords power management
servers
Language English
License Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org
LinkModel DirectLink
MeetingName ISCA '11: The 38th Annual International Symposium on Computer Architecture
MergedId FETCHMERGED-LOGICAL-a247t-3e65517fba8855debc7f5f83a6bae9da52fb534a5f979ef6c5524109c3e63ab3
PageCount 12
ParticipantIDs acm_books_10_1145_2000064_2000103_brief
acm_books_10_1145_2000064_2000103
ieee_primary_6307768
PublicationCentury 2000
PublicationDate 20110604
2011-June
PublicationDateYYYYMMDD 2011-06-04
2011-06-01
PublicationDate_xml – month: 06
  year: 2011
  text: 20110604
  day: 04
PublicationDecade 2010
PublicationPlace New York, NY, USA
PublicationPlace_xml – name: New York, NY, USA
PublicationSeriesTitle ACM Conferences
PublicationTitle 2011 38th Annual International Symposium on Computer Architecture (ISCA)
PublicationTitleAbbrev ISCA
PublicationYear 2011
Publisher ACM
IEEE
Publisher_xml – name: ACM
– name: IEEE
SSID ssj0000974452
ssj0019956
Score 2.3365307
Snippet Much of the success of the Internet services model can be attributed to the popularity of a class of workloads that we call Online Data-Intensive (OLDI)...
SourceID ieee
acm
SourceType Publisher
StartPage 319
SubjectTerms Abstracts
Computer systems organization -- Architectures -- Distributed architectures -- Client-server architectures
Information systems -- Data management systems -- Middleware for databases -- Application servers
Information systems -- Data management systems -- Middleware for databases -- Database web servers
Servers
Title Power management of online data-intensive services
URI https://ieeexplore.ieee.org/document/6307768
WOSCitedRecordID wos000292709800028&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB7a4qGnqq1YX0QQvBjbbp57FosHKT0U6W1Jsgn00Fb6-v0m2XRFEMTT7oZdWIZMvslkvm8AHgzxuwpBckyo9yZashLrUTnCwhBVCkO5jNyqj3cxmcj5PJ824KnmwlhrY_GZfQ638Sy_XJt9SJUNOAniM7IJTX-puFp1PmXoA2PKaumowDyOzCIPwZjLIP_XHlEWDgJ9OHnUekrPPGn--IFBFlfwmG0JLRACaJnlj9YrEXnGnf_98yn0vil8aFqD0xk07OocOsceDii5dBeyaWiThpZ1FQxaO1TJZ6BQPYoXxxp3tE3LSg9m49fZyxtOfRSwyqjYYWK5j4uE00pKxkqrjXDMSaK4VjYvFcucZoQq5nKRW8cNYx7Xh7nxHxKlyQW0VuuVvQTkt658qIkHOB9GBWjLjfd36ZzItCSU9eHeW6sI-4NtUVGeWZEsWiSL9uHxz3cKvVlY14dusGfxWeluFMmUV78PX0O7yveGDMkNtHabvb2FE3PYLbabuzhZvgDm1rDJ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEB1qFeypaivWzwiCF9e2m2STPYulYi09FOktJNkEemgr_fD3m6TpiiCIt92wC8uwkzczmfcG4E5jl1UwnCeYOG8iBS0S1S26CdNYFkyTjAdu1fuADYd8MslHFXgouTDGmNB8Zh79ZTjLLxZ640tl7Qx78Rm-B_t-clZka5UVlY4LjQktxaM89zhwixwIJxn3AoC1LqH-KNAFlDu1p3ifRdUft9BOwx4e6i1-CIKHLT37MXwlYE-v_r-vPoLmN4kPjUp4OoaKmZ9AfTfFAUWnbkA68oPS0Kzsg0ELi7YCGsj3jybTXZc7WsWNpQnj3vP4qZ_ESQqJTAlbJ9hkLjJiVknOKS2M0sxSy7HMlDR5IWlqFcVEUpuz3NhMU-qQvZNr9yKWCp9Cdb6YmzNALnnNOgo7iHOBlAe3XDuP59ayVHFMaAtunbWEzxBWYkt6piJaVESLtuD-z2eEWk6NbUHD21N8bJU3RDTl-e_LN3DYH78NxOBl-HoBtW3119dLLqG6Xm7MFRzoz_V0tbwOP84XWOy0Eg
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2011+38th+Annual+International+Symposium+on+Computer+Architecture+%28ISCA%29&rft.atitle=Power+management+of+online+data-intensive+services&rft.au=Meisner%2C+D.&rft.au=Sadler%2C+C.+M.&rft.au=Barroso%2C+L.+A.&rft.au=Weber%2C+W.&rft.date=2011-06-01&rft.pub=IEEE&rft.isbn=9781450304726&rft.issn=1063-6897&rft.spage=319&rft.epage=330&rft_id=info:doi/10.1145%2F2000064.2000103&rft.externalDocID=6307768
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1063-6897&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1063-6897&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1063-6897&client=summon