|> home > about > news|
All users are reminded that the HPCx service will close 31st Jan 2010. Both the main service and the archive will close on 31st January. Users should plan to move off/copy alldata that they require prior to this. Any data which remains on the system after this time will be lost.The secure disposal of all data and decommissioning of the servers commences on Monday 1st February.
As such, we would recommend that users start the process of retrieving the data they require as soon as possible (if they have not already done so). We would strongly advise against waiting until the last minute to do this.
Transferring large amounts of data to/from HPCx - bbFTP:
If you have large amounts of data to transfer to or from HPCx, we can recommend the use of bbFTP. We have installed bbFTP on HPCx (and also on HECToR) in /usr/local/packages/bbftp. Usage instructions are available in the HPCx user guide at:http://www.hpcx.ac.uk/support/documentation/UserGuide/HPCxuser/Getting_started.html
The most efficient way to transfer data from HPCx to HECToR is via GridFTP. If your group have certificates then we would recommend this is the method you use. As mentioned above, bbFTP is also installed on HECToR. If you don't have certificates then we would recommend you use this method.
The SAFE (Service Administration Software) will remain available for a period of three months following the end of service. This will enable PIs to extract any usage data as necessary to complete any final reports.
Moving to HECToR:
Any PIs who wish to move compute resource from HPCx to an existing HECToR project should contact the helpdesk. PIs who do not currently hold a project on HECToR, but who wish to set one up should contact the helpdesk as soon as possible. A HECToR technical assessment will be required, however we will provide disk space immediately to enable you to transfer files.
DEISA Extreme Computing Initiative Call for Proposals now open, Submission Deadline: 16th February 2010
The DEISA Extreme Computing Initiative, or DECI, is an excellent way to gain access to a large amount of HPC cycles, for projects which are too large to run on a single National Resource.
Last year's call saw eight UK DECI projects granted a total of over 10 million hours on a variety of platforms spread across Europe. Indeed, the DECI not only gives you cycles but man power is also available to help you exploit the DEISA infrastructure.
DEISA, or the Distributed European Infrastructure for Supercomputing Applications, consists of 11 HPC Centres across Europe, providing a pan-European Grid of HPC resources. HPCx has been a platform within DEISA for the last 4 1/2 years and HECToR joined DEISA in May, 2008, although it is worth noting that UK projects will be assigned machines based outwith the UK.
The DECI is a scheme through which European computational scientists can apply for single-project access to world-leading computational resources in the European HPC infrastructure, operated by DEISA, for a period of up to 10 months. DECI aims to enable European researchers to obtain access to the most powerful national computing resources in Europe, regardless of their country of origin or work and to enhance DEISA's impact on European science and technology at the highest level.
Projects supported by DECI will be chosen on the basis of innovation potential, scientific excellence and relevance criteria. Priority will be given to proposals that promote collaborative research, either at a cross- national or cross-disciplinary level. Further, proposals from PIs that have yet to benefit from DECI compute and applications enabling resources may be given preference. Proposals selected under this call will be given access to the infrastructure for applications enabling from 1st July 2010 and for production runs from 1st October 2010 to 30th April 2011.
The following applications profiles are particularly suited to the DECI:
DEISA can provide applications enabling support (typically up to several months) for the enabling and optimisation of these applications.
For more information, please visit:
Following on from the success of the EPSRC Open Access Initiative on HPCx, a new initiative for Low Priority Access has been implemented.
In addition to the existing "paid" queues, a set of "free" low priority queues have been introduced. These queues are designed to:
When a paid job is submitted, it may have to wait until a free job finishes, but it is ensured by (1) that the wait time will be minimal. As ensured by (2), the paid job will then run ahead of any waiting free jobs, and can utilise any of the available paid queues. Capability use of the system is encouraged by (3).
The following six low priority queues have been added.
To route jobs into these queues you need to add the following directive in your loadleveler script
#@requirements = ( Feature == "low_priority" )
Your job will then run in one of the above parn queues. This can be checked using the llq command.
The new queues will be available after today's maintenance on HPCx. This initiative will then run as a trial over the next couple of months. If you have any questions on this, please do not hesitate to contact the helpdesk.
|http://www.hpcx.ac.uk/about/news/||contact email - email@example.com||© UoE HPCX Ltd|