
Sustainability & AI
Saving the Planet, One Algorithm at a Time! You all know I love high-performance computing, but recently I’ve been thinking
Home » Manchester Metropolitan University
Manchester Metropolitan University (MMU), a leading UK university recognised for its multi-disciplinary research and commitment to innovation, sought to expand its high-performance computing capabilities to meet the growing demands of data-intensive research, foster cross-departmental collaboration, and empower students and faculty to pursue cutting-edge projects.
Despite its established reputation in academia, the institution had had never operated a centralised, shared HPC available for use by all faculties. Seeking to supercharge existing and future research projects, ranging from data-intensive analytics to simulations in engineering and the sciences, university leaders recognised that a robust HPC environment was critical for driving new discoveries and academic excellence.
While the MMU team had a wealth of technical experience, it did not have the in-house capability or time required to build an HPC initiative from the ground up. Furthermore, the team would have had to have mastered traditional HPC concepts and the intricacies of cloud-based resource provisioning quickly, when time was already extremely pressured.
In addition, multiple HPC systems were scattered across departments, which offered little continuity and introduced unnecessary operational risks.
Matthew Jones, Senior Linux Infrastructure Engineer at MMU, said: “Due to the tight time frame and the strict requirement to ensure security and compliance, the project would have required a great deal of internal resource if we’d done it in house.
“The live environment had to be up and running for internal users by September. Indeed, with only a few months to go, planning, configuration, and testing needed to happen at pace – leaving little room for error.
“As a higher education institution, we also needed strict adherence to governance and compliance standards for data protection. This meant we had to ensure sensitive research data was properly secured within the new cloud-based HPC environment.
“In summary, every aspect of the HPC project, technical, procedural and organisational, represented new territory for the university and we recognised that even the smallest miscalculation in HPC design or deployment could have led to spiralling costs, missed deadlines or even compromised research outputs.”
Recognising the urgent need for a reliable and multifunctional HPC system, MMU needed to call in the experts on HPC and cloud architecture – Red Oak Consulting who we knew had a solid track record and reputation within the research and tertiary education sector.
Red Oak’s consultants started by conducting a series of workshops and interviews with the technical teams to clarify performance targets, data security needs, user requirements, and likely usage patterns. Armed with this knowledge, they designed for us an architecture blueprint detailing compute, storage, networking, and job scheduling design in Microsoft Azure.
The pilot system is being moved to a new, segmented Azure tenancy, allowing the HPC environment to be managed separately from other university IT resources, reducing risks, streamlining governance, and enabling clearer cost tracking.
One of the biggest considerations at the time was MMU’s unfamiliarity with in-depth HPC, Red Oak providing hands-on training sessions and bespoke documentation, guiding it through the nuances of HPC clustering, workload orchestration, performance tuning, and cost monitoring. This knowledge transfer was vital for ensuring that, once live, the system could be operated and refined in-house. Before opening the HPC resource to the wider university, Red Oak’s consulting team helped establish a testing phase to validate real workloads. From stress-testing the cluster’s ability to handle large parallel jobs, to confirming data security configurations, Red Oak’s systematic approach gave the project leads confidence that the HPC environment would perform robustly under production conditions.
Beyond the initial go-live, Red Oak also offered strategic advice on scaling and potential future enhancements. This included recommendations on data lifecycle management, advanced visualisation capabilities, and integrating emerging technologies such as GPUs and AI. By anticipating tomorrow’s needs, the university can continue to evolve their HPC environment in a controlled and cost-effective manner.
It’s been a pleasure working with the MMU team on this digital transformation. Working in a true collaborative manner, the progress their team has made is very impressive. The new HPC environment offers a scalable, secure platform that will empower researchers and drive innovation across the university.”
Owen Thomas, Red Oak Consulting
As the university gets ready to go live, the transformation is already evident. Through targeted guidance, the MMU IT team has gone from novices to competent HPC practitioners. They now have a stronger grasp of cloud-based HPC architecture and feel equipped to support researchers across the institution.
The new, integrated HPC system ensures standardisation, robust data protection, and centralised management, eliminating the previous scattered approach and associated vulnerabilities. Researchers now stand to benefit from secure, streamlined access to powerful compute resources, fostering greater efficiency and collaboration across departments.
Furthermore, by separating the HPC environment into its own Azure tenancy, the university gains the flexibility to scale resources up and down as needed. This avoids the capital costs associated with on-premises systems while ensuring HPC can be quickly adapted to the evolving research landscape.
The new HPC environment, once fully operational, will help the university attract bigger research grants and more ambitious scientific projects. It represents a competitive edge in fields where computational power can significantly accelerate innovation and discoveries.
Thanks to Red Oak’s in-depth consultation, the university has a clear roadmap for the platform’s evolution and this pioneering HPC initiative is set to open new horizons in research and academic collaboration across the university.
Matthew Jones
Manchester Metropolitan University
Sign up to our newsletter to stay up to date with all the latest news and advancements in High-Performance Computing
Saving the Planet, One Algorithm at a Time! You all know I love high-performance computing, but recently I’ve been thinking
The Accessible World of HPC Supercomputers sure do grab headlines and solve some of the world’s largest problems, but being
Introduction It’s mid-2013 and I have just graduated with a physics degree. I really enjoyed the theoretical component, but the
Tech Skills in Action at Sheffield Hallam Yesterday, my colleague James Page and I supported the final stage of a
36 Pure Offices
Cheltenham Office Park
Hatherley Lane
Cheltenham GL51 6SH
United Kingdom
Telephone: +44 (0)1242 806 188
General Enquiries
info@redoakconsulting.co.uk
Sales Enquiries
sales@redoakconsulting.co.uk
© Copyright Red Oak Consulting 2024 | Privacy Policy | Modern Slavery Policy
Get in touch with our team of HPC experts to find out how we can help you with your HPC, AI & Cloud Computing requirements across:
Call us on
+44 (0)1242 806 188
Experts available:
9:00am – 5:30pm GMT
"*" indicates required fields