Infrastructure

Back to NaN

Burns & Smithers (aka Vespignani & Menczer)

Burns & Smithers (aka Vespignani & Menczer)

Computing and Storage Infrastructure

Thanks to funding from NSF (Award No. IIS-0811994) and the School of Informatics, NaN and other CNetS researchers under the supervision of PIs Fil Menczer and Alex Vespignani can avail themselves of an advanced computing and storage infrastructure.

The infrastructure is composed of two servers (Burns and Smithers) each with 16 cores and large amounts of memory (32GB and 128GB respectively), and a 12TB disk array, all integrated into a fibre-channel GFS Storage Area Network. This infrastructure is particularly amenable to analyzing and mining large network data.

FAQ

How do I get an account?

Given the source of funding, this infrastructure is reserved for certain projects managed by PIs Menczer and Vespignani. PIs must approve requests. Once a request is approved, it can be submitted to our support team.

How do I submit technical requests (ie software installations) to the support team?

Use the CSG Help Desk.

What are the names of the servers? How do I connect? Which server should I use?

burns.cs.indiana.edu and smithers.cs.indiana.edu. They run Linux so you connect via ssh. To be allowed to login, your IU account must be added to the cnets group. Always use Burns for your computing jobs unless you need more than 32GB of memory. Burns is by far the less busy, and therefore the faster of the two machines because Smithers runs many Web and database services.

What disk/file system should I use?

Go into /l/cnets/research/ (/l/cnets is the SAN/GFS shared disk array) and create a directory there with your username, and only use that directory. Nothing will be backed up. Please remove unneeded files. The SAN is to store data we currently use, not for long-term backup. Please use MDSS for long-term backup of your data.

Can I email other users of the infrastructure?

Yes but first you have to prove that you are a human.

Whom should I thank?

If you use this infrastructure to support your research, be sure to acknowledge NSF Award No. IIS-0811994 in your papers.

What else should I know?

Please be respectful of others when running your jobs:

  • always use nice to share CPU fairly, e.g. nice nohup program &
  • use the commands du and df to check disk usage
  • use vmstat to make sure your job does not thrash memory (even if we have a lot)
  • use man to learn about any commands you’re not familar with, e.g. man nice, man du, man vmstat…
  • google ‘unix tutorial’ to learn about process, disk, and memory management, then if you’re still lost, ask someone…
  • you can use ssh to do a simple and secure login to the server, here is a brief tutorial for how to set up ssh.

Happy computing!