MSc Teaching Cluster care and feeding

Note that this page is currently being developed contact iainr@infREMOVE_THIS.ed.ac.uk for more information.

This page covers the design and maintenance of the Msc Teaching cluster

The subclusters

There are actually two groups of machines:

  • One cluster consists of machines which support GPGPU computing, primarily through the CUDA framework
  • The other cluster supports more tradition parallel computing though hadoop and other software frameworks including gridengine. Note that in order for gridengine to work ssh access is currently blocked on port 22 apart from a select number of hosts infrastructure hosts defined in the tcpwrappers stansa in mscteach_node.h. ssh is not blocked on scutter1 and scutter2 as these are hadoop submit nodes.

The gridengine scheduler is on forthea and the head nodes are porthemmet and tatties (in the forum server room).

The remainder of the machines are in the MSC racking in the Appleton Tower.

For more details of configuring software etc.

See CareandFeeding-CDTCluster and HadoopCareAndFeeding

The cluster is set up in a similar manner as documented there.

-- IainRae - 15 Jun 2016

Topic revision: r6 - 05 Jul 2017 - 15:27:37 - GrahamDutton
 
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback
This Wiki uses Cookies