资源管理软件TORQUE与作业调度软件Maui的安装、设置及使用

Similar documents
PowerPoint 簡報

Microsoft Word - linux命令及建议.doc

Windows 2000 Server for T100

ebook 185-6

Microsoft Word - PS2_linux_guide_cn.doc

Cadence SPB 15.2 VOICE Cadence SPB 15.2 PC Cadence 3 (1) CD1 1of 2 (2) CD2 2of 2 (3) CD3 Concept HDL 1of 1

ebook140-9

Windows 2000 Server for T100

untitled

網路安全:理論與實務 第二版

幻灯片 1

录 环 录结 统 资 查 环 设 设 环 变 库 问题 业 计 结 资 请 业 业查 WRF CESM

投影片 1

untitled

IP505SM_manual_cn.doc

5 whuTrainee

ebook

Guide to Install SATA Hard Disks

Serial ATA ( Silicon Image SiI3114)...2 (1) SATA... 2 (2) B I O S S A T A... 3 (3) RAID BIOS RAID... 5 (4) S A T A... 8 (5) S A T A... 10

IP Access Lists IP Access Lists IP Access Lists

AL-MX200 Series

Basic System Administration

WebSphere Studio Application Developer IBM Portal Toolkit... 2/21 1. WebSphere Portal Portal WebSphere Application Server stopserver.bat -configfile..

05_資源分享-NFS及NIS.doc

本文由筱驀釹贡献

Some experiences in working with Madagascar: installa7on & development Tengfei Wang, Peng Zou Tongji university

ebook 132-6

软件测试(TA07)第一学期考试

epub83-1

WinMDI 28

epub 73-5

untitled

2004 Sun Microsystems, Inc Network Circle, Santa Clara, CA U.S.A. Sun Sun Berkeley BSD University of California UNIX X/Open Company, Ltd.

投影片 1

Red Flag Linux Desktop 4.0 Red Flag Linux Desktop 4.0 1

自由軟體教學平台

投影片 1

1 o o o CPU o o o o o SQL Server 2005 o CPU o o o o o SQL Server o Microsoft SQL Server 2005

P4i45GL_GV-R50-CN.p65

雲端 Cloud Computing 技術指南 運算 應用 平台與架構 10/04/15 11:55:46 INFO 10/04/15 11:55:53 INFO 10/04/15 11:55:56 INFO 10/04/15 11:56:05 INFO 10/04/15 11:56:07 INFO

PowerPoint Presentation

Oracle 4

電子商業伺服器管理(終極版).doc

BlackBerry Classic Smartphone-用户指南

Pchome

untitled

スライド 1

ebook70-22

HCD0174_2008

2004 Sun Microsystems, Inc Network Circle, Santa Clara, CA U.S.A. Sun Sun Berkeley BSD UNIX X/Open Company, Ltd. / SunSun MicrosystemsSun

1 WLAN 接 入 配 置 本 文 中 的 AP 指 的 是 LA3616 无 线 网 关 1.1 WLAN 接 入 简 介 WLAN 接 入 为 用 户 提 供 接 入 网 络 的 服 务 无 线 服 务 的 骨 干 网 通 常 使 用 有 线 电 缆 作 为 线 路 连 接 安 置 在 固 定

ebook140-11

南京航空航天大学CPU/GPU集群使用手册

ebook62-1

Serial ATA ( nvidia nforce4 Ultra/SLI)...2 (1) SATA... 2 (2) B I O S S A T A... 3 (3) RAID BIOS RAID... 6 (4) S A T A... 9 (5) S A T A (6) Micro

1 SQL Server 2005 SQL Server Microsoft Windows Server 2003NTFS NTFS SQL Server 2000 Randy Dyess DBA SQL Server SQL Server DBA SQL Server SQL Se

untitled

2 2 3 DLight CPU I/O DLight Oracle Solaris (DTrace) C/C++ Solaris DLight DTrace DLight DLight DLight C C++ Fortran CPU I/O DLight AM

K7VT2_QIG_v3

自由軟體教學平台

Abstract arm linux tool-chain root NET-Start! 2

User ID 150 Password - User ID 150 Password Mon- Cam-- Invalid Terminal Mode No User Terminal Mode No User Mon- Cam-- 2

自由軟體教學平台

ebook140-8

TCA Linux 相容性認證測試流程步驟

PowerPoint 演示文稿

AL-M200 Series

考 試 日 期 :2016/04/24 教 室 名 稱 :602 電 腦 教 室 考 試 時 間 :09: 二 技 企 管 一 胡 宗 兒 中 文 輸 入 四 技 企 四 甲 林 姿 瑄 中 文 輸 入 二 技 企 管 一

Microsoft Word - Front cover_white.doc

スライド 1

谚语阐因

1.ai

社大規畫-生活藝能期末報告.doc

ebook

1 重 要 提 示 基 金 管 理 人 的 董 事 会 及 董 事 保 证 本 报 告 所 载 资 料 不 存 在 虚 假 记 载 误 导 性 陈 述 或 重 大 遗 漏, 并 对 其 内 容 的 真 实 性 准 确 性 和 完 整 性 承 担 个 别 及 连 带 责 任 基 金 托 管 人 中 国

NT 4

untitled

ebook35-2

HOL-CHG-1695

Chapter 2

工银瑞信货币市场证券投资基金2008年度第2季度报告

untitled

ansoft_setup21.doc

$$% % $ (%) % %$ $ ( *+,)(-)-./0-1//0- %) %) % - $%2)33%0 $ % ((3./. 3/3 )3 / % (()33(1 % (()3(/ %89856%:;< % (()3 0()0 3 (. <<=330(<</ 3 3. ()

Microsoft Word - 在VMWare-5.5+RedHat-9下建立本机QTopia-2.1.1虚拟平台a.doc

Ác Åé å Serial ATA ( Sil3132) S A T A (1) SATA (2) BIOS SATA (3)* RAID BIOS RAID (4) SATA (5) SATA (a) S A T A ( S A T A R A I D ) (b) (c) Windows XP

CANVIO_AEROCAST_CS_EN.indd

untitled

Sun Update Connection System Sun Microsystems, Inc Network Circle Santa Clara, CA U.S.A

epub 61-2

深圳市亚可信息技术有限公司 NetWeaver 7.3 EhP1 ABAP on Redhat Enterprise Linux Server 62 for Oracle112 High Availability System Installation Created by

Ch. 2

<4D F736F F D20EEA3BDDDB7FECEF1C6F7CCD7BCFED3C3BBA7CAD6B2E156332E302E646F63>

EK-STM32F

.. 3 N

Serial ATA ( Nvidia nforce430)...2 (1) SATA... 2 (2) B I O S S A T A... 3 (3) RAID BIOS RAID... 6 (4) S A T A... 9 (5) S A T A (6) Microsoft Win

ebook70-5

目 录 Linux Mint 简介... 3 Linux Mint 安装... 6 Linux Mint 桌面初识 软件管理...30 小技巧...40 总结...42

一 Grass 是 什 么 1 简 介 GRASS (Geographic Resources Analysis Support System, 地 理 资 源 分 析 支 持 系 统 ) 是 最 负 盛 名 的 开 源 地 理 信 息 系 统 (GIS) 以 下 是 它 的 一 些 特 点 : 1

Transcription:

TORQUE Maui hmli@ustc.edu.cn 2008 1 1 TORQUE 2 1.1 TORQUE........................... 2 1.2 TORQUE...................... 2 1.3 TORQUE.......................... 4 1.4 TORQUE........................... 4 2 Maui 5 2.1 Maui............................ 5 2.2 Maui............................ 5 3 6 3.1................................... 7 3.2................................... 8 3.3.............................. 8 3.3.1 qstat.................... 9 3.3.2 qhold........................... 10 3.3.3 qrls............................ 10 1

3.3.4 qdel canceljob..................... 10 3.3.5 checkjob....................... 11 3.3.6 qorder.................. 12 3.3.7 qselect............ 12 3.3.8 showq................... 13 3.3.9 pbsnodes qnodes................. 13 2

1 TORQUE TORQUE Maui http://www.clusterresources.com TORQUE http://www.clusterresources.com/torquedocs21/ Maui http://www.clusterresources.com/products/maui/docs/ mauiusers.shtml 1.1 TORQUE kd50 node0101 root@kd50# tar zxvf torque-2.2.1.tar.gz root@kd50# cd torque-2.2.1 root@kd50#./configure prefix=/opt/torque-2.2.1 with-rcp=rcp with-rcp=rcp rsh withrcp=scp scp rcp scp root@kd50# make root@kd50# make install 1.2 TORQUE TORQUE /etc/profile TORQUE=/opt/torque 2.2.1 MAUI=/opt/maui 3.2.6p20 if [ `id u` eq 0 ]; then PATH= /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin: PATH=$PATH:$TORQUE/bin:$TORQUE/sbin:$MAUI/bin:$MAUI/sbin else PATH= /usr/local/bin:/usr/bin:/bin:/usr/games:$torque/bin:$maui/bin PATH=$PATH:$TORQUE/bin:$MAUI/bin fi Maui Maui Maui 3

source /etc/profile root TORQUE root@kd50#./torque setup root /var/spool/torque/server priv/nodes kd50 node0101 node0101 node0101 np=2 /var/spool/torque spool undelivered drwxrwxrwt chmod 1777 spool undelivered root@kd50# pbs server -t create root@kd50# qmgr Qmgr: dque Qmgr: create queue dque queue type=execution Qmgr: set server default queue=dque Qmgr: set queue dque started=true Qmgr: set queue dque enabled=true Qmgr: set server scheduling=true pbs server pbs server # shutdown server qterm t quick # start server pbs server # verify all queues are properly configured qstat q # view additional server configuration qmgr c 'p s' # verify all nodes are correctly reporting pbsnodes a # submit a basic job echo sleep 30 qsub 4

# verify jobs display qstat 1.3 TORQUE TORQUE root@kd50# make packages torque-package-clients-linux-i686.sh torque-package-devel-linux-i686.sh torque-package-doc-linux-i686.sh torque-package-mom-linux-i686.sh torque-package-server-linux-i686.sh root@node0101#./torque-package-clients-linux-i686.sh install 1.4 TORQUE /var/spool/torque TORQUE server name NFS TORQUE /var/spool/torque/mom priv/config $pbsserver kd50 # note: hostname running pbs server $logevent 255 # bitmap of which events to log $usecp kd50:/home /home $pbsserver $usecp home 5

/etc/profile TORQUE=/opt/torque 2.2.1 if [ `id u` eq 0 ]; then PATH= /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin: PATH=$PATH:$TORQUE/bin:$TORQUE/sbin else PATH= /usr/local/bin:/usr/bin:/bin:/usr/games PATH=$PATH:$TORQUE/bin fi source /etc/profile pbs mom 2 Maui TORQUE pbs sched Maui Maui 2.1 Maui root@kd50# tar zxvf maui-3.2.6p20-snap.1182974819.tar.gz root@kd50# cd maui-3.2.6p20 root@kd50#./configure prefix=/opt/maui-3.2.6p20 with-pbs=/opt/torque- 2.2.1 root@kd50# make root@kd50# make install 2.2 Maui /usr/local/maui/maui.cfg SERVERHOST kd50 # primary admin must be first in list ADMIN1 root # Resource Manager Definition RMCFG[KD50] TYPE=PBS@RMNMHOST@ 6

RMTYPE[0] PBS /etc/profile TORQUE=/opt/torque 2.2.1 MAUI=/opt/maui 3.2.6p20 if [ `id u` eq 0 ]; then PATH= /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin: PATH=$PATH:$TORQUE/bin:$TORQUE/sbin:$MAUI/bin:$MAUI/sbin else PATH= /usr/local/bin:/usr/bin:/bin:/usr/games PATH=$PATH:$TORQUE/bin:$MAUI/bin fi source /etc/profile Maui root@kd50# maui pbs sched 3 TORQUE Maui qsub TORQUE Maui qsub TORQUE 7

3.1 serial job.pbs #!/bin/sh #PBS N job name #PBS o job.log #PBS e job.err #PBS q dque cd yourworkdir echo Running on hosts `hostname` echo Time is `date` echo Directory is $PWD echo This job runs on the following nodes: cat $PBS NODEFILE echo This job has allocated 1 node./yourprog 1 TORQUE PBS PBS #PBS qsub yourworkdir dque job name job.log job.err #PBS -N -o -e q job name job.log job.err dque user@kd50: /work$ qsub ser job.pbs 3 7. kd50 37.kd50 37 kd50 1 `hostaname` ` 8

3.2 #!/bin/sh #PBS N job name #PBS o job.log #PBS e job.err #PBS q dque #PBS l nodes=4 cd yourworkdir echo Time is `date` echo Directory is $PWD echo This job runs on the following nodes: cat $PBS NODEFILE NPROCS=`wc l<$pbs NODEFILE` echo This job has allocated $NPROCS nodes mpiexec machinefile $PBS NODEFILE np $NPROCS./yourprog #PBS -l nodes= mpiexec user@kd50: /work$ qsub par job.pbs 3.3 TORQUE Maui canceljob checkjob nqs2pbs nqs pbs pbsnodes printjob qdel 9

qhold qmove qnodes pbsnodes qorder qrls qselect qstat qsub showbf showq showstart tracejob TORQUE Maui 3.3.1 qstat qstat user@kd50: /work$ qstat Job i d Name User Time Use S Queue 4 8. kd50 job name4 user 0 E dque 4 9. kd50 job name1 user 0 0 : 0 0 : 0 0 R dque 5 0. kd50 job name2 user 0 H dque 5 1. kd50 job name3 user 0 Q dque E Q H R 10

3.3.2 qhold qhold qstat H 50.kd50 user@kd50: /work$ qhold 50.kd50 3.3.3 qrls qrls user@kd50: /work$ qrls 50.kd50 3.3.4 qdel canceljob qdel canceljob user@kd50: $ qdel 50.kd50 user@kd50: $ canceljob 51.kd50 11

3.3.5 checkjob checkjob user@kd50: $ checkjob 51.kd50 checking job 51 State : Hold Creds : user : user group : user c l a s s : dque qos :DEFAULT WallTime : 0 0 : 0 0 : 0 0 o f 9 9 : 2 3 : 5 9 : 5 9 SubmitTime : Sun Dec 2 1 9 : 2 2 : 1 9 ( Time Queued Total : 0 0 : 4 6 : 1 3 E l i g i b l e : 0 0 : 2 4 : 4 0 ) Total Tasks : 4 Req [ 0 ] TaskCount : 4 P a r t i t i o n : ALL Network : [NONE] Memory >= 0 Disk >= 0 Swap >= 0 Opsys : [NONE] Arch : [NONE] Features : [NONE] IWD: [NONE] Executable : [NONE] Bypass : 0 StartCount : 0 PartitionMask : [ALL] Flags : RESTARTABLE PE: 4.00 S t a r t P r i o r i t y : 24 cannot s e l e c t job 51 f o r p a r t i t i o n DEFAULT ( non i d l e s t a t e ' Hold ' ) State: Hold user@kd50: $ checkjob 49.kd50 checking job 49 State : Running Creds : user : user group : user c l a s s : dque qos :DEFAULT WallTime : 1 : 0 7 : 1 4 o f 9 9 : 2 3 : 5 9 : 5 9 SubmitTime : Sun Dec 2 1 9 : 0 2 : 1 0 ( Time Queued Total : 0 0 : 0 0 : 0 1 E l i g i b l e : 0 0 : 0 0 : 0 1 ) StartTime : Sun Dec 2 1 9 : 0 2 : 1 1 Total Tasks : 4 Req [ 0 ] TaskCount : 4 P a r t i t i o n : DEFAULT Network : [NONE] Memory >= 0 Disk >= 0 Swap >= 0 Opsys : [NONE] Arch : [NONE] Features : [NONE] 12

NodeCount : 4 A l l o c a t e d Nodes : [ node04 : 1 ] [ node03 : 1 ] [ node02 : 1 ] [ node01 : 1 ] IWD: [NONE] Executable : [NONE] Bypass : 0 StartCount : 1 PartitionMask : [ALL] Flags : RESTARTABLE Reservation '49 ' ( 1:06:52 > 9 9 : 2 2 : 5 3 : 0 7 Duration : 9 9 : 2 3 : 5 9 : 5 9 ) PE: 4.00 S t a r t P r i o r i t y : 1 State: Running 3.3.6 qorder qorder user@kd50: $ qstat Job i d Name User Time Use S Queue 5 2. kd50 job name1 user 0 H dque 5 3. kd50 job name2 user 0 Q dque 5 4. kd50 job name3 user 0 Q dque user@kd50: $ qorder 53.kd50 54.kd50 user@kd50: $ qstat Job i d Name User Time Use S Queue 5 2. kd50 job name1 user 0 H dque 5 4. kd50 job name3 user 0 Q dque 5 3. kd50 job name2 user 0 Q dque qorder 53.kd50 54.kd50 53.kd50 54.kd50 54.kd50 53.kd50 3.3.7 qselect qselect 13

5 2. kd50 user@kd50: $ qselect -s H 3.3.8 showq user@kd50: $ showq ACTIVE JOBS JOBNAME USERNAME STATE PROC REMAINING STARTTIME 52 user Running 4 9 9 : 2 2 : 4 4 : 0 9 Sun Dec 2 2 1 : 0 4 : 3 7 1 Active Job 4 o f 4 P r o c e s s o r s Active (100.00%) IDLE JOBS JOBNAME USERNAME STATE PROC WCLIMIT QUEUETIME 54 user I d l e 4 9 9 : 2 3 : 5 9 : 5 9 Sun Dec 2 2 1 : 0 4 : 4 5 1 I d l e Job BLOCKED JOBS JOBNAME USERNAME STATE PROC WCLIMIT QUEUETIME 53 user Hold 4 9 9 : 2 3 : 5 9 : 5 9 Sun Dec 2 2 1 : 0 4 : 3 7 Total Jobs : 3 Active Jobs : 1 I d l e Jobs : 1 Blocked Jobs : 1 3.3.9 pbsnodes qnodes pbsnodes qnodes free down offline user@kd50: $ pbsnodes -l free node0101 f r e e node0102 f r e e node0104 f r e e 14