Technical Perspective: THE DATA CENTER IS THE COMPUTER.
The article discusses MapReduce, a scheme developed by Google for managing the workloads for data center computers. A data center is apt of have a wide variety of machines with differing capacities. MapReduce addresses this by making assignments based on how efficiently a machine has completed recen...
| Publicado en: | Communications of the ACM Vol. 51; no. 1; pp. 105 - 106 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
Association for Computing Machinery
Jan2008
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=28106661&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 28106661 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 00010782 ACM jtl: Communications of the ACM issn: 00010782 maglogo: N pubinfo: dt: Jan2008 vid: 51 iid: 1 pid: 68 pub: Association for Computing Machinery artinfo: ui: 28106661 10.1145/1327452.1327491 ppf: 105 ppct: 1 formats: tig: atl: Technical Perspective: THE DATA CENTER IS THE COMPUTER. aug: au: Patterson, David A. affil: Pardee Professor of Computer Science, U. C. Berkeley. Fellow and Past President of ACM. su: Data libraries Distributed computing Google Inc. Computer networks Electronic data processing Information storage & retrieval systems sug: subj: Data libraries Distributed computing Google Inc. Computer networks Electronic data processing Information storage & retrieval systems ab: The article discusses MapReduce, a scheme developed by Google for managing the workloads for data center computers. A data center is apt of have a wide variety of machines with differing capacities. MapReduce addresses this by making assignments based on how efficiently a machine has completed recent work. Particularly slow machines will have incomplete workloads reassigned to faster machines. The program also makes allowances for hardware failure. MapReduce is simple enough for even novice programmer to use. pubtype: Periodical doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y dt: @attributes: year: 2008 holdings: @attributes: islocal: N |
|---|