Csdcf

content for csdcf people

SAIL

Group Contact Information:

Andrew Ng: ang@cs.stanford.edu
Alex Sandra Pindedo: asandra@cs.stanford.edu

Stanford AI Lab Group Notes

Printing

Printing in SAIL is entirely centralized on the linux systems -- they should not have any local printer configuration at all, but should be automatically picking up the queues on robo.stanford.edu, which acts as the print server.

Psychology

NOTE: Effective FY 12-13 only support Psychology Department

Group Contact Information:

Bernadette Connolly: bconnolly@stanford.edu

SCSNL

Stanford Cognitive and Systems Neuroscience Lab

And

FIND Lab

Location

Servers are in Gates B14, rack D7.

Group Contact Info

Psychiatry fMRI Group -- Brain Imaging

Head: Vinod Menon : menon@stanford.edu
Financial Admin: Joanna Shuxian Yu: joannayu@stanford.edu

Locations:

  • 780 Welch Road, Suite 201 (people, printers, PCs). 
  • A an office on Arastradero Rd.

Mike Huang was the old sysadmin (up to 2009) - 6507668777

Servers:

All servers are puppetized, except for the Xserve.

NLP

Group Contact Information:

Chris Manning: manning@stanford.edu
Dan Jurafsk: jurafsky@stanford.edu
Alicia Middleton: amiddlet@stanford.edu

 

CSLI

Group Contact Information:

Stanley Peters : peters@csli.stanford.edu

Amita Kumar: akumar@stanford.edu

 

Old help site is still useful: http://help-csli.stanford.edu/computers/

BIAC

The BIAC machines include three Linux RAID servers racked in Jordan 333:

  biac1 (Dell PowerEdge, with four RAID boxes plus cpu)
  biac2 (ExcelMeridian, with two 16-disk RAID boxes plus cpu)
  biac3 (Western Scientific, with one internal RAID and one external RAID)

two other Linux machines:

  sns2 (NIS/Samba server and tower machine in Jordan 427)
  sns7 (NIS client and old compute server racked in Jordan 333)

and a couple of live Windows XP machines (as well as some dead ones, in Jordan 427):

Bejerano

Group Contact Information:

Gill Bejerano : bejerano@stanford.edu

Hoa Nguyen: hoa@stanford.edu

Please update this document if you use it and know the actual steps

A few policies we'd like to make automated whenever we get a newcomer into the lab:

1. Add them to the group "evodevo". This group has not been generated yet but we'd like all of us to be part of it (rather than "users") so that we can all write to group-wide space.

Backup using rdiff-backup

mbc

The research server machine mbc in the Psychology department runs rdiff-backup from cron each night to update the clone of /home that lives in /back.

rdiff-backup effectively retains incremental backups up in the process of updating the clone, hence the name of the program.

Note that /back is unmounted after the backup run finishes, so that users won't accidentally access it or, worse, delete stuff from it.

Backup using Legato

Legato, aka Sun Networker, is a commercial backup system used in CS on the machine loft.stanford.edu to back up about 20 Windows machines.

Loft has an LTO-2 tape drive and generally writes about 290GB of data on each tape, which generally holds about 15 to 20 days worth of backups.

Backup using Arkeia

Math and Psych

Arkeia is a commercial backup system used by CSD-CF in the Math and Psychology departments.

For Arkeia version 8 or newer, the interface is a web page. On psych and mathfs, it's at port 20617 (note: http, NOT https).

Arkeia is currently used on:

  • mathfs (version 9)
  • psych (verstion 8)

(The interface to Arkeia for versions before 8 is an X11 program called xarkeia.  We no longer have such older versions active.)

Syndicate content