Due to the relative heterogeneity of the SCARF cluster, it can be difficult do determine how busy it is at any particular moment. This information can be important if you need access to a specific part of the cluster or if you are submitting jobs remotely and thus cannot query the queuing system directly. This page is intended to address all of these problems at the same time. Each box represents a single processor core, and each block of boxes represents a different logical part of the cluster. A key to the different statuses is available below.

Please note that the SCARF-LEXICON, SCARF-LEXICON-2 and SCARF-IBIS nodes are not generally open to most users. Also ISIS users and some CLF users have priority access to shares of SCARF 13, 14 and 15. When not in use by those groups the nodes are only available for general use of up to 24 hours.


SCARF is available at all times, with the exception of essential planned maintenance to hardware or software and unexpected failures. Any problems that occour out of hours are dealt with the next working day, and hardware support is undertaken by the vendor on an agreed contract. Please note that Wednesday 8:30am -10:00am should always be considered as an 'at risk' period, as this is when essential maintenance is sheduled to occur. Reliability and performance statistics are available to all users via Ganglia results.


  Free   Reserved   Busy
  Closed   Unavailable
Intel '15 & '16 (-m "scarf15 scarf16") 380 priority cores for ISIS, 400 for LSF Octopus
Intel '17 (-m scarf17) (2112 priority cores for ISIS)