Sponsored Content
Special Forums UNIX and Linux Applications High Performance Computing How to check performance of your HPC cluster? Post 302643935 by otheus on Monday 21st of May 2012 04:02:26 AM
Old 05-21-2012
"Understanding" isn't so important. At the end of the results, there is a number labeled "GLOPS". It looks like this:
Code:
WR00L2C2        8192    16     1     1             327.60          1.119e+00

The first column expresses some of the options used in the options (HPL.dat) file. The next two columns express the size in dimensions. The last column -- the one you want -- is the number of Gigaflops measured for that run. It's in scientific notation, meaning:
Code:
1.119 x 10^(00)  = 1.119 GFlops

If the output were 3.42e-01, then you'd have:
Code:
3.42 x 10^(-01) = 0.342 GFlops = 342 MFlops

Do you understand now?

---------- Post updated at 10:02 AM ---------- Previous update was at 09:34 AM ----------

As to how to configure this to run on multiple nodes, you have to (1) configure HPL.dat to run on mxn processors, where m and n are close together; so if you have a cluster containing 12 cores, optimal is normally m=3 and n=4. That's not always the case, but it typically is. (2) You must configure your MPI environment so that it runs on that many processors. How you do #2 depends on your MPICH software and setup. If you have a batch system like PBS or SGE, that further complicates the matter, as you need to queue the job with that number of processors. Sometimes with queuing engines, you can't easily dictate which nodes the job will run on. So if you have 6 nodes each with 4 cores, it might not be possible to say "run this on 3 nodes using all cores on each one". (But normally it's possible to say "run it on 3 nodes using 1 core on each one".)
These 2 Users Gave Thanks to otheus For This Post:
 

8 More Discussions You Might Find Interesting

1. Solaris

Performance check without a counterpart

Hi, is it possible to check out the speed limits of a box without a remote peer for netio and stuff? I transferred 1 Terrabyte last night, which is not really much and need to find the bottleneck. The remote server is soon going to retire, yet I need to copy the 8T to the new machine. This... (9 Replies)
Discussion started by: PatrickBaer
9 Replies

2. Shell Programming and Scripting

how to check all the applications are in cluster using shell script

Hi I have an application running in four different node.The server is tomcat.Each node in each tomcat server.How do i check whether all the nodes are in cluster using shell script. any command to check this would be of great use.:) (2 Replies)
Discussion started by: ahamed
2 Replies

3. Shell Programming and Scripting

Monitoring script to check if cm cluster is up or not.

hi guys have this little problem, need some help from script gurus. basically I'm running hpux cmviewcl command, cmviewcl command will produce db1pkg up running enabled box1 my script PSV='box1' STAT='up' check_db1pkg() { # assign cmviewcl output... (2 Replies)
Discussion started by: sparcguy
2 Replies

4. UNIX for Dummies Questions & Answers

How to check if the server is on a Cluster

Hi im connecting to a datacenter remotely. is there a command to know if the server is on a cluster? i want to know the command to use in these OS(hp-ux,solaris,linux) Thanks (6 Replies)
Discussion started by: jinslick25
6 Replies

5. Linux

Cluster check

I am working on a Linux server and want to check the cluster status. I dont know how many server is in cluster and what is the command to check. could you please help me to get the cluster status? (4 Replies)
Discussion started by: anshu ranjan
4 Replies

6. AIX

How to check if a filesystem is part of a cluster

Hello, - How do I know if a filesystem is part of a cluster? - Or do I have to check if the vg related to the fs is part of a cluster instead? if so, how do I check it? - I would also need to check if there are vxfs type inside aix machines and if there are, how do I know if that type of... (2 Replies)
Discussion started by: asanchez
2 Replies

7. Red Hat

How to check performance?

Hi, all What would be the a,b,c in troubleshooting slow performance on RH box, I type and it became really slow, what commands or log files to examine. What parameters to check? Thanks all T (2 Replies)
Discussion started by: trento17
2 Replies

8. Solaris

System Check Performance Tuning

Hello Forum, Well I am fairly new to this Solaris os thing. One thing I would like to check for system health and performance. I know the codes like prstat,vmstat,sar,iostat,netstat,prtdiag -v, What else does a want to be sys admin have to look for when checking a solaris box? I know... (3 Replies)
Discussion started by: br1an
3 Replies
pbs_resources_sp2(7B)							PBS						     pbs_resources_sp2(7B)

NAME
pbs_resources_sp2 - pbs resources on IBM SP-2 DESCRIPTION
This page summarizes the resources supported by PBS for the IBM SP-2. These resources are specified by including them in the -l option argument on the qsub or qalter command or in the PBS job script. RESOURCES
nice The nice value under which the job is to be run. Units: unitary. walltime Maximum amount of real time during which the job can be in the running state. Units: time. arch Specifies the administrator defined system architecture requried. This defaults to whatever the PBS_MACH string is set to in "local.mk". Units: string. host Name of host on which job should be run. This resource is provided for use by the site's scheduling policy. The allowable values and effect on job placement is site dependent. Units: string. nodes Number and/or type of nodes to be reserved for exclusive use by the job. The value is one or more node_specs joined with the '+' character, "node_spec[+node_spec...]. Each node_spec is an number of nodes required of the type declared in the node_spec and a name or one or more properity or properities desired for the nodes. The number, the name, and each properity in the node_spec are separated by a colon ':'. If no number is specified, one(1) is assumed. Units: string. The name of a node is its hostname. The properities of nodes are: . ppn=# specifying the number of processors per node requested. Defaults to 1. . arbitrary string assigned by the system administrator, please check with your administrator as to the node names and properities available to you. Examples: . To ask for 12 nodes of any type: -l nodes=12 . To ask for 2 "server" nodes and 14 other nodes (a total of 16): -l nodes=2:server+14 The above consist of two node_specs "2:server" and "14". . To ask for (a) 1 node that is a "server" and has a "hippi" interface, (b) 10 nodes that are not servers, and (c) 3 nodes that have a large amount of memory an have hippi: -l nodes=server:hippi+10:noserver+3:bigmem:hippi . To ask for three nodes by name: -l nodes=b2005+b1803+b1813 . To ask for 2 processors on each of four nodes: -l nodes=4:ppn=2 . To ask for 4 processors on one node: -l nodes=1:ppn=4 . To ask for 2 processors on each of two blue nodes and three processors on one red node: -l nodes=2:blue:ppn=2+red:ppn=3 host Allows a user to specify the desired execution location. This resource is provided for use by the site's scheduling policy. The allowable values and effect on job placement is site dependent. Units: string. other Allows a user to specify site specific information. This resource is provided for use by the site's scheduling policy. The allow- able values and effect on job placement is site dependent. Units: string. software Allows a user to specify software required by the job. This is useful if certain software packages are only available on certain systems in the site. This resource is provided for use by the site's scheduling policy. The allowable values and effect on job placement is site dependent. Units: string. EXAMPLES
qsub -l nodes=15,walltime=2:00:00 script or in a qsub script as a directive: #PBS -l nodes=15,walltime=2:00:00 RESOURCE UNITS
The resource values take the following units: time specifies a maximum time period the resource can be used. Time is expressed in seconds as an integer, or in the form: [[hours:]minutes:]seconds[.milliseconds] unitary The maximum amount of a resource which is expressed as a simple integer. SEE ALSO
the PBS ERS, qsub(1B), qalter(1B), pbs_submit(3B), pbs_alterjob(3B), pbs_statjob(3b) pbs_resources_aix4(7B), pbs_resources_irix5(7B) pbs_resources_sunos4(7B), pbs_resources_unicos8(7B) Local pbs_resources_sp2(7B)
All times are GMT -4. The time now is 10:19 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy