Difference between revisions of "HPC engineering TODO List"

From Define Wiki
Jump to navigation Jump to search
 
(46 intermediate revisions by 2 users not shown)
Line 1: Line 1:
−
== Benchmarking ==
+
== General Stuff ==
−
* Openfoam benchmarking (ARM vs Sandy bridge)#
 
−
** Single node tests
 
−
** OpenMPI for full 12/48 calx nodes
 
−
* Blender, open source rendering. To get a feel for ARM vs Intel on media projects
 
−
* jbench (see what the JVM performance is like on a system, ejre vs openjdk)
 
−
* Itaru (BWA genome stuff) pBWA doesnt scale (needs 12 more systems)
 
−
* justonedb guys, mysql vs postgres vs justonedb with their data sets (bloom analytics software) Peter
 
−
* spec cpu and hepspec on arm
 
−
** '''DP''' to email spec
 
−
* openmpi 1.6 results for all tests
 
−
** '''JH''' IMB tests, pingpong np=2, all tests 2 nodes, 4 nodes, 8 nodes
 
−
* linpack HPL
 
  
−
== Software testing ==
+
* OCP testing
−
* Hadoop terasort on v1.0.3 apache
+
* GPU test Drive
−
* Install / tests CDH (Cloudera distribution of hadoop)
+
* '''Tidy Up!!!!'''
−
* '''JH''' Intel e3-1220L-v2 ivy bridge, sysbench power/performance tests
+
* Scan / Document our System
−
* '''JH''' cxmanage on the headnode, and on wiki
 
−
* VDI and spice for ubuntu / fedora, test on ARM, instructions on wiki
 
−
* Openstack, newest release with dashboard
 
−
* ScaleIO to port software to ARM (source and rebuild)
 
−
* Zettaset
 
−
** '''DP''' put in touch with UK partner #
 
−
* Ubuntu stuff to do:
 
−
** MaaS
 
−
** JuJu
 
−
** Landscape
 
−
** apt-proxy
 
  
−
== Hardware Bits ==
+
== Dave ==
−
* Server range for Viglen
 
−
** Stage 1: Hardware on SB similar to current range
 
−
** Stage 2: Additional storage products for schools (cheap NAS / iSCSI solution) to compete against hitachi
 
−
** Stage 3: Solutions, cloud, vmware?
 
  
−
== Mike TODO List ==
+
== Mike ==
−
* PCM cluster
 
−
** Install intel compilers and licenses (you can take from the headnode ~david/software/intel_compilers -DONE
 
−
** Install the intel cluster check ~david/software/platform/PCM 3.1 -DONE
 
  
−
* K20
+
* Piston Cloud
−
** There must be another network port! - DONE
+
* RHEL / Centos 7 testing
−
** Install CUDA
+
* New Cobbler System
 +
* Open Stack
 +
* Linux Training - 28th January
  
−
** Rocks Cluster
+
== Alex ==
−
*** Get ACML installed in /share/apps/acml/5.x -DONE
 
  
 +
* 10GB Card Testing
 +
* 10GB Blog Post
 +
* 10GB Training - 25th Feb
 +
* Ceph
 +
* Lustre
  
−
* Benchmarks
+
== Mouza ==
−
** Get used to building codes, here are some of the standard ones we use as part of HPC testing
 
−
** HPL (Linpack)
 
−
** Stream (memory bandwidth)
 
−
** IMB (Network throughput and latency)
 
−
** Iozone (disk performance)
 
−
** Give them a try when you get a chance
 
  
−
* ARM Benchmarks
+
* Alinea
−
** Oil and Gas: Seismic processing benchmarks: http://geocomputing.narod.ru/benchmark.html (give a go on an ARM box and intel box)
+
* 8 GPU / PHI system benchmarking
 +
* 2nd Blog
 +
* Arm Hepspec
 +
* Hadoop
  
−
== Completed section ==
+
== Adam ==
 +
 
 +
* Imperial Blade Testing
 +
* Emulex 10GbE Testing
 +
* Unit.tv Lustre

Latest revision as of 12:57, 9 July 2015

General Stuff

  • OCP testing
  • GPU test Drive
  • Tidy Up!!!!
  • Scan / Document our System

Dave

Mike

  • Piston Cloud
  • RHEL / Centos 7 testing
  • New Cobbler System
  • Open Stack
  • Linux Training - 28th January

Alex

  • 10GB Card Testing
  • 10GB Blog Post
  • 10GB Training - 25th Feb
  • Ceph
  • Lustre

Mouza

  • Alinea
  • 8 GPU / PHI system benchmarking
  • 2nd Blog
  • Arm Hepspec
  • Hadoop

Adam

  • Imperial Blade Testing
  • Emulex 10GbE Testing
  • Unit.tv Lustre
Retrieved from "http://wiki.define-technology.com/mediawiki-1.35.0/index.php?title=HPC_engineering_TODO_List&oldid=8611"