Tuesday 27th February 2018

Planned RDF Downtime for UPS Battery Replacement

The RDF has built-in UPS systems (uninterruptible power supply) to help prevent data loss and corruption in case of a power outage. The batteries in these UPS systems are now approaching the end of their life and have to be replaced. The replacement work needs to be carried out with the RDF offline to reduce any risk of data loss during the process and is required to be completed before the actual battery end of life date.

The work is planned for the week of 5 March 2018 and should take between 3 and 5 days to complete. The RDF, DAC, DTNs, GridPP and access to JASMIN will be unavailable from 9am on 5 March 2018 and are planned to be returned to service by 17:00 on 9 March 2018. The /eprsc, /nerc and /general file systems will be unavailable. ARCHER and all other services will be available as usual. Once work commences these timescales will be reviewed and if it is possible the RDF will be returned to service earlier than planned. We will send out an update if plans change and will keep this page up to date.

If you require access to your data on the RDF next week please allow enough time to copy the data to the required location. We recommend not waiting until Friday in case the copying process takes longer than anticipated.

Please contact the helpdesk if you have any questions or concerns.

Modules to be removed during Maintenance Session on 28th February 2018

We are planning to remove old modules on ARCHER to allow for the installation of newer modules. The modules which will be deleted have not been part of a default set previously. We are keeping the two previous default sets and will not be changing the default set at this time.

A full list of modules to be deleted as part of this maintenance is available at

If you have any questions or concerns about these planned removals please contact the ARCHER Helpdesk.

RAP Calls Open for UK HPC Access

Three RAP (Resource Allocation Panel) calls are currently open for access to HPC resources for research projects within EPSRC remit:

  • RAP: Access to ARCHER
  • RAP: Top-up of ARCHER resource on EPSRC grants
  • RAP: Open Access to Tier-2

All calls have the following deadlines:

  • Technical Assessment: submitted to appropriate helpdesk by 22 Mar 2018
  • Call Closing Date: Applications submitted by 10 Apr 2018
  • Panels Meet: May 2017
  • Earliest project start: May 2017
  • Latest project start: Jun 2017

RAP: Access to ARCHER

EPSRC offers access to ARCHER through calls for proposals to the Resource Allocation Panel (RAP). Users can request significant amounts (>1,000kAUs or >66,667 ARCHER core hours) of computing resource over a maximum one-year period. The aim of this call is to provide access to our national state-of-the-art high performance computing facility for proposals of high scientific quality that would benefit from ARCHER.

A non-exclusive list of eligible projects include:

  • Short computational projects that do not warrant a full grant application
  • UK led collaborative projects with international and/or industry partners
  • Joint applications from students (as Co-I) with proven HPC experience (e.g. a successfully completed instant access project) and their PIs
  • Projects that link consecutive standard grant applications or that aid the preparation of a grant or fellowship application
  • Extended feasibility studies and trialling application developments at scale

Further information can be found on the call page:

RAP: Top-up of ARCHER resource on EPSRC grants

Access to the national high performance computing resource can be allocated as part of EPSRC grants at the time of award. Initially time-limited resource are granted and grant holders are invited to apply for a top-up to meet the objectives of their EPSRC grant. Applicants to this call can request ARCHER compute resource only to facilitate the remainder of their existing EPSRC grant.

Further information can be found on the call page:

RAP: Open Access to Tier-2

EPSRC is offering open access to five Tier-2 High Performance Computing facilities through this call for proposals. The five facilities users can access through this call are:

Further details on each of these centres can be found through their websites and in the call document. The aim of this call is to: Provide access to our national Tier-2 HPC facilities for adventurous high-risk, high-reward projects that will benefit from the diversity of computing architectures available at Tier-2.

A non-exclusive list of eligible projects includes:

  • Short computational projects that do not warrant a full grant application;
  • UK led collaborative projects with international and/or industry partners;
  • Joint applications from students (as Co-Is) with proven HPC experience and their PIs;
  • Projects that link consecutive standard grant applications or that aid the preparation of a grant or fellowship application;
  • Extended feasibility studies and trialling application developments at scale;
  • High-risk, high-reward projects that would benefit from using novel architectures.

Further information can be found on the call page:

Exascale Applications and Software Conference (EASC 2018) - Edinburgh, UK. 17-19 April 2018 - Call for posters

Hot Topic Posters: Poster submissions on recent or on-going research in the topics of interest listed at are warmly welcomed. Poster presenters will have the opportunity to present a lightning talk (1-2 minutes) introducing the topic of their poster. Please send a brief abstract (max. 100 words, plus one image) to by 1st of March 2018.

Early bird registration now open

Registration is now open and you will be able to enjoy our early bird discount until the 28th of March 2018. Please visit to register and for more details.

Hands-on Porting and Optimisation Workshop: Making the most of ARCHER

Thursday March 8th, 2018, Oxford e-Research Centre, University of Oxford

This is an excellent opportunity to receive one-on-one guidance and individually tailored advice from high-performance computing specialists about how best to make use of ARCHER. Attendees will be given accounts on ARCHER for the workshop, with dedicated access on the day to significant portions of the machine.

You will be assigned an ARCHER technical advisor in advance, who will work with you directly during and in some cases before the workshop to help you in any way you need.

The workshop is aimed at:

  • Current users of the ARCHER service (including developers) looking to improve the performance of their code, e.g. parallel scaling and/or IO, in order to make more efficient use of the machine.
  • New users of the ARCHER service looking to port existing software - be it their own code or a third-party application - to run on the machine for the first time or to explore its performance prior to an application for time on ARCHER.
  • Potential future users of the service looking to understand whether ARCHER is suitable for their computing needs and could potentially accelerate their research.
  • Users who want to learn more about data management or using the RDF (Research Data Facility) and how this can help their work on ARCHER.

The workshop is open to anyone interested in using ARCHER; it is not restricted to current users. We encourage applications from small teams of collaborators. The workshop is tailored to individual participants and we can cater to the needs of attendees with a wide range of previous experience.

This workshop is free to all. For more information and to register please visit

State of the ART I/O Tools - Virtual Tutorial

Wednesday 28th February 2018 15:00 Online

Elsa Gonsiorowski, Lawrence Livermore National Laboratory

I/O has historically been an afterthought in application development and performance analysis. As processing power increases and accelerators increase performance, Amdahl's law increasingly points to I/O as an opportunity for performance optimization. Unfortunately, the tool framework for analysis of I/O performance is not as robust as that for computational performance. Tools that exist, such as IOR and Darshan, target a very small slice of I/O performance. Additionally complicating this space is the introduction of hierarchical storage into the I/O space. With the increased prevalence of burst buffers and other forms of hierarchical storage, we can expect to find new answers to historically well-studied problems such as optimal checkpointing frequency.

This talk will cover basic I/O strategies and cover state-of-the-art I/O tools that can be used to prepare for burst buffer architectures. It will include MACSio (an I/O benchmark emulating physics-simulations) and the Scalable / Checkpoint Restart (SCR) library.


Upcoming Training Opportunities

Registration open now

  • Porting and Optimisation Workshop Oxford 8 March 2018
  • Efficient Parallel IO on ARCHER Cambridge 20 - 21 March 2018
  • Threaded Programming Southampton 20 - 22 March 2018
  • Message-Passing programming with MPI Southampton 11 - 13 April 2018
  • Advanced OpenMP Cambridge 17 - 19 July 2018
  • Full details and registration at