The Rocket HPC pilot phase is due to end on 31st October and the service will then be considered LIVE. This has been a team effort and we are more than grateful to all the NUIT staff, pilot users and other people who have helped us get to this point.

For projects that have not yet registered to use Rocket or that are waiting for their registration to be completed, there will be no change: we are completing registrations when we can and most delays are related to software requests. Please bear with us, we know you are waiting.

For projects that have already been added to Rocket, the move to a live service means:

– project PIs will be able to add more members to their group and instructions will be available on http://www.ncl.ac.uk/itservice/research/hpc. Please do not try this just yet – it may not work well.

– the service will no longer be subject to interruptions at short notice.

– queries and requests will be handled as service desk tickets.

As always, most information about Rocket is on the web pages at http://www.ncl.ac.uk/itservice/research/hpc. Please get in touch if you have any another queries.

The Rocket HPC service now has its first pilot users registered on the system and starting to run jobs. Some issues have already come to light and we are working to resolve those. Our thanks go to the pilots for working with us on this.

The pilot phase is intended to be fairly brief and aims to assess the system’s performance as well as unearthing issues. We then expect to start widening access towards the end of the month. To help us plan for this, you may now register projects for the HPC service at http://www.ncl.ac.uk/itservice/research/hpc/hpcregistration/

Registering now will let you tell us about your needs (e.g. for software applications) so that we can plan the work we need to do, and will help speed your access onto the system.

As always, please keep an eye on the HPC web pages at http://www.ncl.ac.uk/itservice/research/hpc for information about the service and how to use it, and get in touch if you have any queries or feedback.

Over the past month, NUIT have worked with our suppliers to configure Rocket and integrate it into the University’s network and systems, and to test its performance. Once NUIT have completed some more of the work needed on our side to manage Rocket as a service, we will be running a short pilot phase with four volunteers who will be helping us iron out any initial problems and assess the system’s performance for a variety of ‘real’ job types.

During the pilot phase we will also start accepting HPC project registrations. When the service is live, all access to Rocket will be controlled by membership of projects, which may be established research projects, taught programs or other groupings – they need not be funded research programmes. Project registration will be via an online form: see http://www.ncl.ac.uk/itservice/research/hpc for more details.

We hope to start widening access to the cluster in September. Watch this space.

It’s been just under a year since the project team were given the go-ahead to start work on building this service. Those of you who attended the requirements-gathering workshops last summer may recall that we intend to launch the service in late Spring/early Summer of this year.

Although the project team is based in NUIT, we have invited input and participation from researchers at every stage of the project: gathering requirements, system architecture, staff recruitment and tender evaluation.  More details will be posted in the coming week or so, but here’s a short summary of the projects progress to date.

People

After a couple of rounds of recruitment we are pleased to be welcoming a new HPC research computing analyst on board. This person will sit at the interface between the technology and the researchers whop use it, providing training, guidance and some coding support.

We are working with a recruitment agency in order to source a systems administrator to oversee the day to day management of the system itself.

The System

Summer’s requirements workshops and questionnaires, were followed up by a day-long solution architecture workshop, where a group of researchers, IT staff and some external HPC expertise evaluated and condensed the various technical requirements into a proposed solution that would serve a wide variety of potential users.

The result was a system based on the most recent generation of Intel Broadwell CPUs, connected via high-speed interconnect, in a topology that allowed for an island of nodes that would favour tightly-coupled parallel jobs. The system design also included groups of medium-, high-, and extra-high RAM nodes, featuring 0.25, .5 and 1.5 TB RAM per node. The 1.5 TB nodes feature 4 CPUs.

Omissions

There are several ways in which our envisaged service falls short of all of the requests for features that we received.Part way through the requiremnts-gathering phase of the project, something called Brexit weighed in. The tender did ask for indicative quotes for GPU nodes, and we have yet to make a final decision on that.  We did not tender for any Intel Xeon Phi nodes, or nodes running Microsoft Windows. We are looking into the possibility of facilitating access to Intel Xeon Phi nodes at the Hartree facility, via the N8 consortium.