Scientific analysis servers

This article describes the scientific  analysis servers. It covers:

  • Analysis  servers 
  • Software installed 
  • Access to storage 
  • When to use it 

Once you have logged on (see above) you can then ssh to the shared analysis machine that you have been given access to. 

Scientific analysis servers 

Scientific analysis servers are designed for interactive and ad-hoc analysis of data in group workspaces, the CEDA archives and users' home directories.  For long running and resource intensive jobs users are required to use  the LOTUS cluster which offers better I/O performance, parallelism and scheduling.

Access to the scientific analysis servers is restricted to ssh login from hosts inside the RAL firewall, so access from outside needs to be achieved via one of the JASMIN login gateway machines (see Table 1).


name of  scientific analysis server 

jasmin-sci1.ceda.ac.uk, jasmin-sci2.ceda.ac.uk, jasmin-sci3.ceda.ac.uk

cems-sci1.cems.rl.ac.uk, cems-sci2.cems.rl.ac.uk


So for example if you had a Jasmin User Login you would type

$ ssh jasmin-sci1.ceda.ac.uk

For more information on using the JASMIN Analysis Platform  click here.

Software installed 

Each analysis server has the following features:

  • Default Redhat Enterprise Linux development environment.
  • The JASMIN Analysis Platform installed, providing commonly-used open source analysis tools.  E.g. NCO, CDO, Python, R.
  • Access to proprietary tools, e.g. IDL and Intel Fortran, through the Module System.
  • Ability to run X-Windows applications by tunnelling your X session through ssh.

For more information on using the Jasmin Analysis Platform   click here.

Access to storage 

Each analysis server has:

  • Access the LOTUS /work/scratch area.
  • Read access to the CEDA Archives according to user permissions.
  • Read/Write access to Group Workspaces according to user permissions.

How and when to use Sci-servers 

 As a JASMIN user of a interactive compute servers the following should be considered:

  • What your process requirements are? ( CPU, memory, execution time should be less that 1 hour otherwise consider batch compute on LOTUS)
  • Be aware of resources available at the time before your process starts and choose the server that is suitable ( check average load on scientific server from message of the day on login server  for example or using any of Linux monitoring command : top, free -h )
  • Monitor your process using top or ps Linux command 
  • Report if there is a user's process affecting the performance of scientific server 

Still need help? Contact Us Contact Us