First, let's start with where you can't run such processes. If you have a CS Sharkestra account, you have access to lots of computers throughout Lindley Hall. However, many of these machines are on the desks of faculty, staff, and students and are not for general use. Just because you are able to login to some random workstation doesn't mean you should be using it. Please get permission before you just randomly pick a Sharkestra machine to use remotely or, better yet, just use the multi-user systems mentioned below.

If you need to run remote processes or a program that will consume a lot of CPU cycles, please try not to let it interfere with other users. You can always run whatever you want to on the machine you are sitting at. However, if you start a CPU intensive process and then logout, leaving it running, it will make that machine much slower for the next person that uses it. You should "nice" the process so it doesn't affect interactive use too much. See the "nice" and "renice" man pages for more information.

The following resources are available for running your remote and CPU-intensive processes:

• burrow.cs.indiana.edu - If you are looking for a system to use in the CS Burrow domain, you can just ssh to burrow.cs.indiana.edu. This will just take you an appropriate burrow system so you don't have to worry about what system to use and have to make changes when systems are upgraded and names change.

• sharks.cs.indiana.edu - If you are looking for a system to use in the CS Sharkestra domain, you can just ssh to sharks.cs.indiana.edu. This will just take you an appropriate Sharkstra system so you don't have to worry about what system to use and have to make changes when systems are upgraded and names change.

• Tank - The multi-user machine named tank.cs.indiana.edu is a Quad Quad-Core (16 total cores) 2.93GHz Intel Xeon system running 32-bit Red Hat Enterprise Linux with 32GB of memory. Tank is in the CS Sharkestra account domain so it is not available to users who only have a CS Burrow account.

• Hulk - The multi-user machine named hulk.cs.indiana.edu is a Quad Quad-Core (16 total cores) 2.93GHz Intel Xeon system running 64-bit Red Hat Enterprise Linux with 32GB of memory. Hulk is in the CS Sharkestra account domain so it is not available to users who only have a CS Burrow account.

• Silo - The multi-user machine named silo.cs.indiana.edu is a 2-processor 2.8GHz Intel Xeon system running Red Hat Enterprise Linux with 8GB of memory. Silo is in the CS Burrow account domain so it is available to users who have a CS Burrow account. Silo is a great place for remote interactive logins and where you should run your server processes. However, this machine is used by a lot of people and tends to be heavily loaded so should NOT be used for CPU-intensive, long-running processes.

• Solaris - There is an alias solaris.cs.indiana.edu that will take you to a Sun system running the Solaris operating system. This is for legacy support of Solaris and you are encouraged to use this system only if you need access to Solaris. Otherwise, using the sharks.cs.indiana.edu and burrow.cs.indiana.edu aliases will take you to faster and more modern systems.

• RI Clusters - The Odin, Sif, Jord, Idun, Tyr, and Woton Clusters were purchased as part of a $2.1M NSF Research Infrastructure grant. These clusters consist of:
  • Sif Cluster - 8 Dual 1.86GHz Intel Quad-Core Xeon systems

  • Odin Cluster - 128 Dual 2.0 GHz AMD Dual-Core Opteron systems

  • Tyr Cluster - 16 Dual 2.0 GHz AMD Dual-Core Opteron systems

  • Jord SMP System - 1 Quad 2.8 GHz AMD Dual-Core Opteron system with 64GB of memory

  • Idun SMP System - 1 Quad 2.3 GHz AMD Quad-Core Opteron system with 16GB of memory

  • Woton Cluster - 16 Dual 2.3 GHz Apple Xserve G5 systems

  Please see the RI Cluster Introduction FAQ and the RI Cluster Network FAQ for more information.

  Access to these clusters is granted by sponsorship of any of the faculty principle investigators (PIs) on the grant. Requests for accounts should be requested by the sponsoring faculty member by emailing accounts-admin.

• Sharkestra - If you are doing experimentation that requires the use of more machines than are available using the above systems then you can use the machines in the open graduate student lab (LH125) if you have a CS Sharkestra account. Just keep in mind that there may be people using these machines at the console so try to use those machines in a way that won't interfere with interactive use. In other words, limit use to jobs that are not memory intensive and nice your processes. In order to get a list of the computers in this lab, just go to the hardware database listing for LH125. There are various Linux workstations in that lab.

• Burrow - If you have only a Burrow account, or also want to use the Burrow machines, then you can use any of the machines in LH004 or LH035. To find the names of the machines in these labs, just go to the hardware database listing for LH004 and LH035. As with the machines in LH125, keep in mind that there may be people using these machines at the console so try to use those machines in a way that won't interfere with interactive use. In particular, limit use to jobs that are not memory intensive and nice your processes.

Also note that there are quite significant computing resources available through UITS. See the Research System Knowledge Base Page for more information.

[Return to the FAQ index]