1、 ETSI TS 102 827 V1.1.1 (2008-08)Technical Specification GRID;Grid Component Model (GCM);GCM Interoperability DeploymentETSI ETSI TS 102 827 V1.1.1 (2008-08) 2 Reference DTS/GRID-0004-1 Keywords interoperability, service, architecture, network ETSI 650 Route des Lucioles F-06921 Sophia Antipolis Ced
2、ex - FRANCE Tel.: +33 4 92 94 42 00 Fax: +33 4 93 65 47 16 Siret N 348 623 562 00017 - NAF 742 C Association but non lucratif enregistre la Sous-Prfecture de Grasse (06) N 7803/88 Important notice Individual copies of the present document can be downloaded from: http:/www.etsi.org The present docume
3、nt may be made available in more than one electronic version or in print. In any case of existing or perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF). In case of dispute, the reference shall be the printing on ETSI printers of the PD
4、F version kept on a specific network drive within ETSI Secretariat. Users of the present document should be aware that the document may be subject to revision or change of status. Information on the current status of this and other ETSI documents is available at http:/portal.etsi.org/tb/status/statu
5、s.asp If you find errors in the present document, please send your comment to one of the following services: http:/portal.etsi.org/chaircor/ETSI_support.asp Copyright Notification No part may be reproduced except as authorized by written permission. The copyright and the foregoing restriction extend
6、 to reproduction in all media. European Telecommunications Standards Institute 2008. All rights reserved. DECTTM, PLUGTESTSTM, UMTSTM, TIPHONTM, the TIPHON logo and the ETSI logo are Trade Marks of ETSI registered for the benefit of its Members. 3GPPTM is a Trade Mark of ETSI registered for the bene
7、fit of its Members and of the 3GPP Organizational Partners. ETSI ETSI TS 102 827 V1.1.1 (2008-08) 3 Contents Intellectual Property Rights6 Foreword.6 Introduction 6 1 Scope 7 2 References 7 2.1 Normative references .7 2.2 Informative references7 3 Definitions and abbreviations.8 3.1 Definitions8 3.2
8、 Abbreviations .10 4 Overall resource specification10 4.1 Principles10 4.2 Overall structure of a descriptor.11 4.2.1 Environment element12 4.2.2 Resources element 12 4.2.3 Acquisition element and desktop grid.13 4.2.4 Infrastructure element .14 5 Infrastructure specification.14 5.1 Common types14 5
9、.1.1 Host list.14 5.1.2 ID15 5.1.3 Path element and path string.15 5.1.4 Argument list 15 5.2 Elements of the infrastructure 15 5.2.1 Host description15 5.2.2 Bridge description.16 5.2.3 Group description .17 6 Instances of bridge protocols (one to one) .18 6.1 rsh.18 6.2 ssh.18 6.3 oarsh .19 6.4 gs
10、issh19 7 Instances of group protocols (One to N) 19 7.1 Direct groups 19 7.1.1 GroupSSH.20 7.1.2 GroupRSH 20 7.1.3 GroupOARSH.20 7.1.4 ARC (NorduGrid).21 7.1.5 CGSP China Grid .22 7.1.6 EGEE gLITE 22 7.1.7 Fura/InnerGrid 23 7.1.8 GLOBUS 23 7.1.9 GridBus.24 7.1.10 IBM Load Leveler 24 7.1.11 LSF .25 7
11、.1.12 Microsoft CCS 26 7.1.13 OAR26 7.1.14 PBS / Torque.27 7.1.15 PRUN28 7.1.16 Sun Grid Engine .29 8 XML files examples (informative).29 ETSI ETSI TS 102 827 V1.1.1 (2008-08) 4 8.1 Bridge protocols .29 8.1.1 rsh .29 8.1.2 ssh .30 8.1.3 oarsh30 8.1.4 gsissh 30 8.2 Group protocols30 8.2.1 GroupSSH.30
12、 8.2.2 GroupRSH 30 8.2.3 GroupOARSH.30 8.2.4 ARC (Nordugrid)30 8.2.5 CGSP China Grid .30 8.2.6 EGEE gLite.30 8.2.7 Fura/InnerGrid 31 8.2.8 GLOBUS 31 8.2.9 GridBus.31 8.2.10 IBM load leveler.31 8.2.11 LSF .31 8.2.12 Microsoft CCS 31 8.2.13 OAR31 8.2.14 PBS / Torque.32 8.2.15 PRUN32 8.2.16 Sun Grid En
13、gine .32 Annex A (normative): XML Schema33 A.1 Common types33 A.2 Main schema 35 A.3 Bridges .39 A.3.1 rsh.39 A.3.2 ssh.39 A.3.3 gsissh39 A.3.4 oarsh .40 A.4 Groups 40 A.4.1 GroupSSH 40 A.4.2 GroupRSH41 A.4.3 GroupOARSH 41 A.4.4 ARC (NorduGrid).41 A.4.5 CGSP China Grid .42 A.4.6 EGEE gLite 42 A.4.7
14、Fura / InnerGrid43 A.4.8 GLOBUS44 A.4.9 GridBus 44 A.4.10 IBM load leveler.44 A.4.11 LSF.45 A.4.12 Microsoft CCS46 A.4.13 OAR .46 A.4.14 PBS / Torque 47 A.4.15 PRUN .48 A.4.16 Sun Grid Engine .48 Annex B (informative): Examples of deployment descriptor.50 B.1 Example 1.50 B.2 Example 2.50 Annex C (n
15、ormative): Custom bridge and group types .52 Annex D (informative): Symbols used in the schema diagrams.54 ETSI ETSI TS 102 827 V1.1.1 (2008-08) 5 Annex E (informative): Bibliography.57 History 58 ETSI ETSI TS 102 827 V1.1.1 (2008-08) 6 Intellectual Property Rights IPRs essential or potentially esse
16、ntial to the present document may have been declared to ETSI. The information pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found in ETSI SR 000 314: “Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notif
17、ied to ETSI in respect of ETSI standards“, which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web server (http:/webapp.etsi.org/IPR/home.asp). Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee can
18、 be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web server) which are, or may be, or may become, essential to the present document. Foreword This Technical Specification (TS) has been produced by ETSI Technical Committee GRID (GRID). The pres
19、ent document is related to document 102-828 (GCM Interoperability Application Description). Introduction The purpose of the present document is to offer a standard, uniform way of describing grid resources, so as to make these resources available to grid-based applications. When building a grid, one
20、 is confronted to several problems, among which: choosing a grid framework, as many are available. when running a grid-enabled application, one must rewrite configuration directives when switching to a different grid framework. a grid can be built on widely different hardware infrastructures, from a
21、n heterogenous set of desktop machines to a dedicated cluster, or even any combination of those. grid infrastructures can also be very dynamic, it should be easy to add or remove machines, and such changes should be transparent to users. To simplify this, the GCM Interoperability Deployment standard
22、 offers a uniform way to describe grid resources, how they are connected to one another, how they can be activated, and how they can be accessed. A grid infrastructure is modelized with several entities: Host, Bridge, and Group (see definitions clause). By combining these entities, one can describe
23、the physical infrastructure of the grid, i.e. how the machines it includes are configured and networked together. The GCM has been first defined in the NoE CoreGRID (42 institutions). A reference Open Source implementation has been tested in the 4 previous GRID Plugtests organized from 2004 to 2007
24、by ETSI, and will also be part of the forthcoming 5thone in 2008. The GridCOMP EU project (FP6, started June 2006 to November 2008) is working to further assess and experiment with the specification. ETSI ETSI TS 102 827 V1.1.1 (2008-08) 7 1 Scope The present document describes an XML schema for des
25、cribing resources to be used in distributed and parallel infrastructures like enterprise and scientific grids and job schedulers. The standard will help enterprises and laboratories to manage large-scale computer and telecom infrastructures with the necessary virtualization. Its primary audience are
26、 grid system administrators who will need to write deployment descriptors for their grid. 2 References References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For a specific reference, subsequent revisions do not apply. Non-specific
27、 reference may be made only to a complete document or a part thereof and only in the following cases: - if it is accepted that it will be possible to use all future changes of the referenced document for the purposes of the referring document; - for informative references. Referenced documents which
28、 are not found to be publicly available in the expected location might be found at http:/docbox.etsi.org/Reference. For online referenced documents, information sufficient to identify and locate the source shall be provided. Preferably, the primary source of the referenced document should be cited,
29、in order to ensure traceability. Furthermore, the reference should, as far as possible, remain valid for the expected life of the document. The reference shall include the method of access to the referenced document and the full network address, with the same punctuation and use of upper case and lo
30、wer case letters. NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee their long term validity. 2.1 Normative references The following referenced documents are indispensable for the application of the present document. For dated references,
31、only the edition cited applies. For non-specific references, the latest edition of the referenced document (including any amendments) applies. 1 ETSI TS 102 828: “GRID; Grid Component Model (GCM); GCM Application Description“. 2.2 Informative references The following referenced documents are not ess
32、ential to the use of the present document but they assist the user with regard to a particular subject area. For non-specific references, the latest version of the referenced document (including any amendments) applies. i.1 ARC / Nordugrid: “middleware for grid computing“. NOTE: Available at http:/w
33、ww.nordugrid.org/. i.2 CGSP / China Grid: “middleware for ChinaGrid“. NOTE: Available at http:/grid.org ETSI ETSI TS 102 827 V1.1.1 (2008-08) 8 i.3 EGEE gLite: “middle-ware for grid computing“. NOTE: Available at http:/glite.web.cern.ch/glite/ . i.4 Fura / Inner Grid: “middleware for grid computing“
34、. NOTE: Available at http:/ i.5 Globus: “open source software toolkit to build Grid systems and applications“. NOTE: Available at http:/www.globus.org/. i.6 GridBus: “middleware for grid computing“. NOTE: Available at http:/www.gridbus.org/middleware/. i.7 IBM Load Leveler: “job scheduler“. NOTE: Av
35、ailable at http:/ http:/ i.8 LSF: “job scheduler“. NOTE: Available at http:/ LSF /. i.9 OAR: “job scheduler“. NOTE: Available at http:/oar.imag.fr/docs/manual.html. i.10 PBS / Torque: “job scheduler“. NOTE: Available at http:/ i.11 PBS / PRUN: “run server from an old version of PBS“. NOTE: Available
36、 at http:/www.cs.vu.nl/das/prun/prun.1.html. i.12 Sun GridEngine: “grid computing framework“. NOTE: Available at http:/ i.13 CCS: “Microsoft Compute Cluster Server“. NOTE: Available at http:/ 3 Definitions and abbreviations 3.1 Definitions For the purposes of the present document, the following term
37、s and definitions apply: bridge: GCM Deployment concept that represents a machine through which multiple machines are accessed NOTE: A typical example is a cluster frontend (see frontend). certificate: electronic document which incorporates a digital signature to bind together a public key with an i
38、dentity - information such as the name of a person or an organization, their address, and so forth certificate authority: entity which issues digital certificates for use by other parties cluster: set of loosely-coupled machines working together so that they are seen as a single computer ETSI ETSI T
39、S 102 827 V1.1.1 (2008-08) 9 Core (CPU Core): processing unit within a processor chip, effectively seen as a CPU from the Operating System NOTE: For instance, a dual-core CPU will be seen as two CPUs by the OS. deployment descriptor: XML file following the schema specified in the present document, w
40、hich describes what grid resources are to be deployed and how frontend: in a physical network, a single machine acting as a gateway for a set of machines NOTE: This is often the case for machine clusters, where individual machines are not accessible directly. Grid Component Model (GCM): component ar
41、chitecture and framework for the portability, and reusability of Grid applications, the integration with SOA infrastructures, and providing interoperability of GRID systems group: GCM Deployment concept that represents a set of machines (i.e. hosts, see definition) and the grid protocol through whic
42、h they can be accessed NOTE: These machines may all be sharing the same configuration (OS type, installation path of executable etc.), but it may also only have the same grid protocol while having very different configurations. A group is generally used to represent a cluster. host: GCM Deployment c
43、oncept which represents a single physical machine on a network NOTE: A GCM Deployment model can be seen as a tree, and hosts are the leaves. Host Capacity (HC): number of programs or VM per Host NOTE: An indicator of the number of programs to be executed in parallel on a host. infrastructure: organi
44、zed set of processing resources which form a grid job scheduler: program which purpose is to manage a queue of program execution requests (jobs) in order to run them in turn, or at a specified time node: GCM Deployment concept that represents a location to execute a computation NOTE: At execution, a
45、 node is a place where a program or an activity can be executed. This can be a single physical machine (a host), or, in the case of a multi-processor / multi-core machine, a single processor or a single core within a machine. resource: processing unit which is a building part of the grid NOTE: A res
46、ource can either be a host (a single physical machine), a bridge (a frontend machine providing access to several others), or a group (a set of identical machines). Service Oriented Architecture (SOA): computer systems architectural style for creating and using business processes, packaged as service
47、s, throughout their lifecycle Schema (XML Schema): set of structure constraints for an XML document NOTE: They provide a means for defining the structure, content and semantics of XML documents. Schemas are a W3C recommendation. See http:/www.w3.org/XML/Schema. Virtual Machine (VM): software impleme
48、ntation of a computer that executes programs like a real machine VM Capacity (VMC): number of activities or node- per VM - an indicator of the number of activities (threads) to be executed in parallel within a program NOTE: In a standard approach, the *Host Capacity x VM Capacity* should be set to the number of CPU cores of the machine. However, these indicators are by no means absolute. They can be tuned to take into account the specifications of the platform
