Service Status

Login Nodes: Available

Last Updated: 06:00, Sunday 25 June 2017


Compute Nodes: Available

Last Updated: 06:00, Sunday 25 June 2017


Low Priority: Disabled

Last Updated: 05:45, Sunday 25 June 2017


This page contains details of the status and usage of the ARCHER system.

Known Issues

No current issues

Open Issues

Open issues
hme04 is currently unavailble on the rdf. Our systems team are working to resolve the issue.

Recently Resolved Issues

Friday 2 June 2017
There was an issue with fs2 between around 20:00 1st June and 04:00 2nd June which led to the failure of some running jobs.
Our Cray team were able to resolve the fs2 problem and restart affected hardware.
We believe that everything is now back in service and Cray are investigating the underlying cause.
Any jobs which failed due to the issue will be refunded.
We apologise for the inconvenience caused.

Upcoming Maintenance Sessions

RDF : Data Analytic Cluster (DAC) and Data Transfer Nodes (DTN) Maintenance
The DAC and DTNs will be unavailable on Monday 26th June 0900 - 1700 BST.
The rdf filesystems (/epsrc, /nerc, /general) will be available via ARCHER.
ARCHER Maintenance
The next full maintenance session is confirmed for Wednesday 28th June 0900 - 1700 BST.
ARCHER login and compute nodes will be unavailable. The rdf will be available via the Data Analytic Cluster (DAC) and the Data Transfer Nodes (DTNs).

Live Usage/Queue Data

Total Usage

Usage history for:

This shows the number of nodes currently in use on ARCHER.

ARCHER usage graph.

Queue Status

Queue history for:

The graph below shows the status of the standard queue on ARCHER.

ARCHER standard queue status graph.

Live Usage/Queue Data by Job Length/Size

This section displays ARCHER usage and queue information broken down by the job class (class is based on size in nodes and run time in hours).

Queue Times by Job Size/Length

The heatmap below provides information on the mean queue times on ARCHER over past periods for different job sizes/lengths. The colour in the heatmap indicates the mean queue time as measured by scheduling coefficient for each class and the number in the box is the number of jobs of that class.

The scheduling coefficient is the ratio of run time to run time plus queue time and, as such, represents the fraction of the time in the queue that was spent running. A value of 1 indicates that the job started instantaneously and a value of 0.5 indicates that the job queued for the same amount of time that it ran for.

Usage history for:

ARCHER scheduling coefficient.

Usage by Job Size/Length

The heatmap below provides information on the usage on ARCHER over past periods for different job sizes/lengths. The colour in the heatmap indicates the number of kAU expended for each class and the number in the box is the number of jobs of that class.

Usage history for:

ARCHER usage by job size/length.

ARCHER Disk Status

RAID Check, fs2: Not Running

Last Updated: 05:45, Sunday 25 June 2017


RAID Check, fs3: Not Running

Last Updated: 05:45, Sunday 25 June 2017


RAID Check, fs4: Not Running

Last Updated: 05:45, Sunday 25 June 2017


ARCHER Network Traffic

Data transferred on and off ARCHER in the last 72 hours (green = data transferred off ARCHER; blue = data transferred on to ARCHER).

ARCHER network traffic graph.

Data transferred on and off RDF in the last 72 hours (green = data transferred off RDF; blue = data transferred on to RDF).

ARCHER network traffic graph.