Algorithms and Software for Emerging Architectures (ASEArch)

Lead Research Organisation: University of Oxford
Department Name: Oxford e-Research Centre


Advanced computing is an essential tool in addressing scientific problems of national interest, including climate change, nanoscience, the virtual human, new materials, next-generation power sources and astrophysics; as importantly it is equally essential to solve commercial and industrial problems in financial modelling, engineering, and real-time decision systems. In recent years the computer systems that underpin these software applications have changed radically and it is no longer possible to simply run the same application software efficiently on new machines. This CCP will address this challenge by creating new knowledge regarding new hardware and software systems and disseminating to other computational science groups. That dissemination will occur through publications, workshops, in-depth study groups, new algorithms and software kernels.

The CCP is building on, and leveraging, several existing networks and institutes and is well positioned to leverage international efforts too. There is agreed engagement with the application-driven CCPs and a broad range of network collaborators.

The activity will have impact on academic computational scientist through direct engagement, availability of technology watch outputs; reports on best practice; papers on new algorithms; and open source software. Through the engagement of our industry partners and appropriate knowledge transfer network we will also have impact on commercial applications.

Planned Impact

The immediate beneficiaries of this initiative will include scientists and developers in the other CCPs, in other computational science projects and in industrial R&D. The impact will be on both the academic and commercial worlds through knowledge transfer in both directions, adoption of software kernels and know-how in industry, through industry input to the technology watch, and dissemination of that into the academic community.

The outputs of ASEArch will be fourfold:

1. An internationally competitive activity resulting in algorithms and software that will lead to more effective use of high-performance chips and computers.
2. Publications and papers on new algorithms, software tools and best practice.
3. An increase in education and training resulting in a new generation of researchers and a more highly skilled workforce.
4. Open source software that implements new algorithms for advanced computing available to anyone for incorporation into applications, products or services.

The activities of this CCP will ensure that the existing investments in scientific software are not lost as new machines architectures become prevalent. The return on investment in future research software and on investments in new HPC hardware will be increased by this initiative.


10 25 50
publication icon
Adámek K (2020) GPU Fast Convolution via the Overlap-and-Save Method in Shared Memory in ACM Transactions on Architecture and Code Optimization

publication icon
Giles M (2016) Algorithm 955 Approximation of the Inverse Poisson Cumulative Distribution Function in ACM Transactions on Mathematical Software

publication icon
Giles MB (2014) Trends in high-performance computing for engineering calculations. in Philosophical transactions. Series A, Mathematical, physical, and engineering sciences

publication icon
Hogg J (2013) A Fast Dense Triangular Solve in CUDA in SIAM Journal on Scientific Computing

publication icon
Hogg J (2016) A Sparse Symmetric Indefinite Direct Solver for GPU Architectures in ACM Transactions on Mathematical Software

publication icon
László E (2016) Manycore Algorithms for Batch Scalar and Block Tridiagonal Solvers in ACM Transactions on Mathematical Software

Description This CCP (Collaborative Computational Project) supported other CCPs in the exploitation of novel architectures. This was accomplished through workshops, training and the development of open source software with widespread applicability.
Exploitation Route The software was equally applicable in both academic and non-academic contexts. Open source software was developed for the use of other CPPs, and some was also contributed to NVIDIA's CUBLAS and CUSPARSE libraries.
Sectors Digital/Communication/Information Technologies (including Software)

Description During the project the other CCPs (Computational Collaborative Projects) accessed our information and findings from our website. Later, we contributed software to NVIDIA's CUBLAS and CUSPARSE mathematical libraries, as well as making software openly available under an open source license at, This research on GPU computing was also fundamental in building our understanding of the power on GPU computing, which in turn led to later activities such as our leadership in creating JADE, the UK's first supercomputer for AI / machine learning:
First Year Of Impact 2013
Sector Aerospace, Defence and Marine,Chemicals,Digital/Communication/Information Technologies (including Software),Education,Energy,Financial Services, and Management Consultancy,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology
Impact Types Economic

Description Future-proof massively-parallel execution of multi-block applications
Amount £282,844 (GBP)
Funding ID EP/K038494/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 08/2013 
End 08/2016
Description JADE: Joint Academic Data science Endeavour
Amount £3,000,000 (GBP)
Funding ID EP/P020275/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 09/2016 
End 03/2020
Description e-Infrastructure South Consortium - Centre for Innovation
Amount £2,820,000 (GBP)
Funding ID EP/K000136/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 01/2011 
End 03/2012
Description CUDA Programming on NVIDIA GPUs 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact One week course on CUDA programming on NVIDIA GPUs, available to both academics and non-academics.

Lots of the students have since gone on to use CUDA programming in their research.
Year(s) Of Engagement Activity 2008,2009,2010,2011,2012,2013,2014