October 03, 2012
Oct. 3 — Data processing, data management and data storage provider, OCF plc, has massively expanded its HPC On-Demand service, enCORE. It can now deliver up to 8,000 cores of vital processing power to a wide range of business sectors in the UK and beyond.
The expansion of the service follows a newly signed agreement between OCF and the Science and Technology Facilities Council's (STFC) Hartree Centre. The Centre is a research collaboratory in association with IBM launched in 2012, which was formed as a result of a £37.5 million investment by the UK government.
The enCORE service will use additional processing power from The Hartree Centre's "Blue Wonder," a new IBM System x iDataPlex cluster comprising 8,192 Intel Xeon E5-2670 processor cores. Tests show that the Blue Wonder iDataPlex cluster can achieve 206.3 teraFLOPS. Its 48 TB shared memory capacity also makes it the largest shared memory cluster in the UK. BlueWonder was installed and configured by OCF in partnership with IBM.
First launched in November 2010, the enCORE service was the first in the UK to operate with a commercial organisation harnessing available processing power from academic and research-based high performance server clusters. The service has since been used successfully by firms such as Engys, Actiflow, CVIS, BHR and Renuda to meet on-going and temporary "burst" requirements for additional processing power.
"enCORE has enabled our clients to re-evaluate their HPC computing strategy. Ease of use, exceptional technical support, scalability and price / performance have been key factors in them deciding to use an off-premise HPC cluster. We've seen this service directly lead to firms winning contracts they could not otherwise have delivered," says Jerry Dixon, HPC on Demand business development manager, OCF plc. "Our expansion of the enCORE service will now enable larger businesses with significant and complex HPC workloads to utilise this flexible facility, and to deliver tangible business benefits."
Dr David Kelsall, senior consultant at fluid engineering consultancy, BHR Group comments: "OCF's enCORE service has enabled us to cope with peaks in demand for capacity when we are undertaking simultaneous consultancy and research projects. It is a very easy service to use, with an uncomplicated and simple structure that doesn't require any previous HPC knowledge to operate."
He adds: "The enCORE service allows us to tackle much larger calculations, up to four times greater than we can manage in-house. As a result, we aren't constrained from taking on more projects than our in-house computing resources allow. It also helps to free up the time of our engineers, who can work on other projects with the extra capacity provided by OCF and can expect a quicker turnaround of analyses results with the enCORE service. We derive a lot of comfort that OCF is providing a UK based 'cloud service', so we know exactly where our data is processed. The enCORE service is cost effective as we can tap into it when required, without the need to tie up a lot of capital for the occasional use of HPC resources."
Zvi Tannenbaum, owner of independent software vendor, Advanced Cluster Systems, says, "OCF's enCORE HPC On-Demand service has acted as a platform enabling us to test and refine our SET (Supercomputing Engine Technology). SET, an MPI (Message Passing Interface) based library, enables mainstream software writers to quickly and cheaply apply MPI parallelization to their software, turning it to a high performance version without code changes, making it suitable for running efficiently on multicore machines and clusters such as OCF's. SET is designed to take the complexity out of MPI parallel programming, making it more readily available to SMEs or other organisations with no in-depth knowledge of MPI.
MPI is the de facto standard for communication among processes in supercomputing centres, and is fully supported by OCF. Now fully tested and operational, OCF is the first HPC On-Demand service provider to offer SET run-time environment in the UK. OCF has demonstrated exceptional technical support and responsiveness during the installation process and continues to deliver professional and timely advice and support."
The enCORE service
· As part of the HPC on Demand service, OCF is responsible for pre-sales qualification with customers to discover required volumes of processing power and benchmarks to demonstrate that enCORE can run specific HPC applications efficiently.
· The service is scalable and suitable for SMEs through to major corporate and academic / research users.
· By working with OCF, customers receive an SLA-driven service, commercial terms and commercial account management, strong technical resource, and first class technical support and assistance for maximum efficiency.
· OCF also holds a number of pre-installed and optimised application codes ready for use with the service; it can work with Independent Software Vendors to get application licensing for the term of a contract with customers, or it can potentially access the end users licences directly, thus ensuring adherence to the ISV's licensing terms.
· enCORE uses the latest Intel and NVIDIA GPU processor hardware for maximum performance.
· Data transfer between the customer and enCORE is handled by enCORE's simple secure web interface or, in the case of extremely large files, by secure shuttle service.
· Contracts with OCF are flexible, and use of enCORE involves a small annual subscription plus a cost per core hour used; interested parties should contact OCF for pricing.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?