THE NUMERICAL ALGORITHMS GROUP LIMITED

Their core work

NAG is an Oxford-based SME that provides numerical and scientific computing libraries, high-performance computing (HPC) expertise, and code optimization services. They help research teams and enterprises squeeze maximum performance from computational software — analyzing bottlenecks, optimizing parallel code, and improving productivity on supercomputing infrastructure. In H2020, they contributed performance analysis tools and HPC optimization expertise to projects tackling computational efficiency and big data challenges in finance.

What they specialise in

How they've shifted over time

NAG's H2020 involvement began in 2015 with two parallel tracks: HPC performance optimization (POP) and big data in financial research (BigDataFinance). By 2018, they doubled down on their core strength by joining POP2, the continuation project with nearly double the funding. The trajectory shows a clear consolidation around HPC performance services rather than diversification into new domains.

NAG is deepening its HPC performance optimization specialization, making them an increasingly focused partner for any project requiring computational efficiency expertise.

How they like to work

NAG exclusively participates as a partner or third party — never as coordinator — which is typical for a specialist SME that contributes deep technical expertise rather than managing large consortia. With 21 unique partners across 10 countries from just 3 projects, they operate within large, well-connected research networks. Their repeat involvement in POP/POP2 suggests they are a trusted, returning partner valued for consistent delivery.

NAG has collaborated with 21 distinct partners across 10 countries through just 3 projects, indicating involvement in large European HPC consortia with broad geographic reach. Their network likely includes major European supercomputing centers and computational research institutions.

What sets them apart

NAG occupies a rare niche as an independent SME specializing in numerical algorithms and HPC performance — most competitors in this space are either academic groups or divisions within large tech companies. Their commercial focus on making scientific code run faster gives them a practical, results-oriented perspective that pure research partners often lack. For any consortium needing to demonstrate computational performance gains, NAG brings both the tools and the track record.

Highlights from their portfolio

• POP2
  Largest funding (EUR 1.49M) and continuation of POP, signaling that the consortium's performance optimization work was successful enough to merit a second phase.
• BigDataFinance
  Marie Curie training network connecting NAG's numerical expertise to financial risk management — an unusual cross-sector application for an HPC specialist.