四章资源管理.ppt
《四章资源管理.ppt》由会员分享,可在线阅读,更多相关《四章资源管理.ppt(68页珍藏版)》请在得力文库 - 分享文档赚钱的网站上搜索。
1、四章资源管理 Still waters run deep.流静水深流静水深,人静心深人静心深 Where there is life,there is hope。有生命必有希望。有生命必有希望Globus与资源规范语言RSLGlobus的资源管理Globus RMSGRAMGRAMGRAMLSFCondorSGEEEApplicationRSLSimple ground RSLInformation ServiceRSLspecializationBrokerGround RSLCo-allocatorQueries&InfoGlobus Components In ActionLocal M
2、achinempirunglobusrunGRAMClientGSIGRAMClientGSIRemote MachineAppNexusAIXPBSMPIgrid-proxy-initX509UserCertUserProxyCertMachines GRAM GatekeeperGSIGRAM Job ManagerGASS ClientRemote MachineAppNexusSolarisUnix ForkMPIGRAM GatekeeperGSIGRAM Job ManagerGASS ClientRSL stringRSL multi-requestRSL single requ
3、estDUROCGASS ServerRSL parserGRAM(Globus Resource Allocation Manager)Overview定位:资源管理的最低层功能:远程运行作业,通过提供的API提交,检测与终止作业GRAM的具体职责处理Resource Specification Language(RSL)形式的作业请求对创建的作业进行远程监控与管理更新MDS的信息Globus Pre-WS Component Interaction DiagramGSIGSIGSIGSIFrom IBM Redbook SG24-6895-012003:Intro to Grid Comp
4、utingGRAM:Grid Resource Allocation Manager GASS:Global Access to Secondary Storage(辅助存储全局访问)MDS:Monitoring and Discovery Service GRIS:Grid Resource Information ServiceGIIS:Grid Index Information ServiceGRAMService that provides remote execution and status management of the requestWhen a job is submi
5、tted by a client,the request is sent to the remote host and handled by the gatekeeper daemon located in the remote host.Then the gatekeeper creates a job manager to start and monitor the job.When the job is finished,the job manager sends the status information back to the client and terminates.GRAM
6、ArchitectureFrom IBM Redbook SG24-6895-012003:Intro to Grid ComputingGRAM ElementsClientsGatekeeper daemon门户监护进程Job ManagerGlobal Access to Secondary Storage(GASS)辅助存储全局访问Dynamically-Updated Request Online Coallocator(DUROC)动态更新请求在线协同分配器User Resource Specification Language(RSL)GRAM ClientsThree clie
7、nts:globusrunglobus-job-runglobus-job-submitGRAM管理流程图示Client APIJob ManagerScheduler Specific PluginJob ProcessGatekeeperJob RequestJob cancelstate change callbackfork/su/execfork/exec/waitspsubmit/spqcondor,lsfgatekeeper的作用gatekeeper:A process,running as root,which begins the process of handling al
8、location requestsperforming mutual authentication of user and resource,determining a local user name for the remote user,starting a job manager which executes as that local user and actually handles the request.In order to start the job manager,the gatekeeper must run as a privileged program相关名词解释Re
9、source An entity capable of running one or more processes on behalf of a userClient The process that is using the resource allocation client-side APIJob A process or set of processes resulting from a job request.Job Request A request to gatekeeper to create one or more job processes,expressed in the
10、 supplied Resource Specification Language.Job Manager One job manager is created by the gatekeeper to fulfill every request submitted to the gatekeeper.GRAM调度与状态转换模型对各个阶段的解释Unsubmitted:The job has not yet been submitted to the schedulerStageIn:The job manager is staging executable,input,or data file
11、s to the jobPending:The job has been submitted to the scheduler,but resources have not yet been allocated for the job.Active:The job has received all of its resources,and the application is executingSuspended:The job has been stopped temporarily by the schedulerStageOut:The job manager is staging ou
12、tput files from the job manager host to remote storage.Done:The job completed successfully.Failed:The job terminated before completion,as a result of an error,or a user or system cancel.GRAM ComponentsGlobus SecurityInfrastructureJob ManagerGRAM client API calls to request resource allocationand pro
13、cess creation.MDS client API callsto locate resourcesQuery current statusof resourceCreateRSL LibraryParseRequestAllocate&create processesProcessProcessProcessMonitor&controlSite boundaryClientMDS:Grid Index Info ServerGatekeeperMDS:Grid Resource Info ServerLocal Resource ManagerMDS client API calls
14、to get resource infoGRAM client API statechange callbacksDUROC(Dynamically-Updated Request Online Co-allocator)Simultaneous allocation of a resource setHandled via optimistic co-allocation based on free nodes or queue predictionadvance reservations will also be supportedglobusrun will co-allocate sp
15、ecific multi-requests using DUROCGRAM ExamplesThe globus-job-run client is a sample GRAM client,using command-line arguments rather than RSL.%globus-job-run pitcairn.mcs.anl.gov/bin/ls%globus-job-run pitcairn.mcs.anl.gov s myprog%globus-job-run pitcairn.mcs.anl.gov s myprog stdin s in.txt stdout s o
16、ut.txt GRAM ExamplesThe globusrun client is a more involved prototype that allows complicated RSL expressions.%globusrun r pitcairn.mcs.anl.gov f myjob.rsl%globusrun r pitcairn.mcs.anl.gov&(executable=myprog)Resource Management APIsGlobus Toolkit has APIs for RSL,GRAM,and DUROC:globus_rslglobus_gram
17、_clientglobus_gram_myjobglobus_duroc_controlglobus_duroc_runtimeResource Specification Language可以用于说明作业要求的通用语言RSL是GRAM的核心部分,它提供了不同组件之间交换信息的手段,比如应用与资源代理之间,资源协同分配与资源管理之间的信息交换形式(attribute=value)需要GRAM理解这些属性attribute Globus提供使用RSL的API可以用于以上之外的更多场合RSL的一些属性(executable=string)Program to runA file path(abso
18、lute or relative)or URL(directory=string)Directory in which to run(default is$HOME)(arguments=arg1 arg2 arg3.)List of string arguments to program(environment=(E1 v1)(E2 v2)List of environment variable name/value pairsRSL的一些属性(stdin=string)Stdin for programA file path(absolute or relative)or URL(stdo
19、ut=string)Stdout for programA file path(absolute or relative)or URL(stderr=string)Stdout for programA file path(absolute or relative)or URL(count=integer)Number of processes to run(default is 1)(hostCount=integer)On SMP multi-computers,number of nodes to distribute the“count”processes across(project
20、=string)Project(account)against which to charge(queue=string)Queue into which to submit jobRSL的一些属性(maxTime=integer)Maximum wall clock or cpu runtime(schedulerss choice)in minutes(maxWallTime=integer)Maximum wall clock runtime in minutes(maxCpuTime=integer)Maximum CPU runtime in minutes(maxMemory=in
21、teger)Maximum amount of memory for each process in megabytes(minMemory=integer)Minimum amount of memory for each process in megabytesRSL Attributes For GRAM(jobType=value)Value is one of“mpi”,“single”,“multiple”,or“condor”mpi:Run the program using“mpirun-np”single:Only run a single instance of the p
22、rogram,and let the program start the other count-1 processes.multiple:Start instances of the program using the appropriate scheduler mechanismcondor:Start a Condor processes running in“standard universe”RSL Attributes for GRAM(gramMyjob=value)Value is one of“collective”,“independent”Defines how the
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 资源管理
限制150内