MICHAEL BANE
[HOME] [Green Software Engineering] [Greener Compute Club]

Research; Publications; Teaching; Support via 'Office Hours'; GSE Portal

I am a Senior Lecturer in Green Software Engineering at Manchester Met University (MMU) and one of the Dept. of Computing & Mathematics' Carbon Literacy Champions.

I have worked in HPC for a few decades, from undertaking research into how to automate parallel programming, and applying HPC to atmospheric aerosol research to running a team supporting researchers to make the best use of HPC. I like to think of HPC in a very board sense (e.g. achieving more, computationally or more efficiently, than whatever currently achieving) and to support others in their HPC journeys.

Computing has a significant environmental cost and my research interests centre around Green Software Engineering, particularly "energy efficient performant computing" (EEPC), looking at ways of undertaking required simulations and analyses but with a lower Carbon footprint. I have recently worked on UKRI-funded "net zero" projects and am a co-author on a £1.9M-funded project's technical report regarding proposals to UKRI for setting out a roadmap for its digital research infrastructure to become net zero by 2030.

Prior to the move to MMU, my teaching previously at the University of Liverpool focussed on HPC, Data Science/Big Data and AI. I was academic lead for the Faculty's "Education Enhancement Group" and a leading member of the Infrastructure subgroup of the university's "Digital" Research Theme. I led on the UG Year in Industry Placements for the Department of Computer Science (2019-2020), and mentored a UG/MSc/PhD student team to 'silver' in the inaugural CI-UK HPC Student Cluster Competition.

I am a Senior Fellow of the HEA, have obtained the Green Software for Practitioners certification (LFC131) with The Linux Foundation, and am also a qualified Mental Health First Aider.

My official MMU/CfACS web page

My Research

My research interests focus around Green Software Engineering with a subfocus on "energy efficient performant computing" (EEPC). Performant computing will remain key to improving science and employing data analytics. However, we need to reduce dramatically the Carbon footprint of perfomant computing. Renewables have one role to play on the supply side, and dramatically improving the energy efficiency of compute has a key role to play on the demand side. Alongside such improvements due to enhanced techniques and technologies, is the need to understand what will motivate stakeholders to more quickly transition to low Carbon solutions, which may include doing less computational experiments (but maintaining progress in science etc).

My current avenues of research include:

...determine which computational regimes it is better to employ accelerators than use traditional CPUs
...how to green AI
...advancing understanding of required aspects of a "green scheduler" (optimising & moving computational workloads, potentially porting to other architectures, application of carbon budgets) and overcoming perceived barriers of moving to a carbon (rather than time or money) focussed approach
...quantifying ICT's environmental footprint within a Digital Sustainable Framework for environmental accounting purposes
...the use of emerging tech (hw) and emerging techniques (sw). In terms of hardware, I am particularly interested in the potential role of FPGA technology and Quantum Computing. On the software side, I am interested in how precision affects time & energy performances;
...being PI of the UKRI Net Zero £120K project "Energy-aware heterogeneous computing at scale" (aka "Energetic"). This project, led by MMU with UCL, EPCC/University of Edinburgh & Newcastle as partners, runs until Dec2022 aims to test whether the use of heterogeneous architecture (use of FPGAs and GPUs alongside CPUs) could significantly reduce energy to solution and thus the energy consumed by UKRI data centres.
...how to better use current hardware and software. For example, we are developing a ML-based tool to better control compiler optimisations and run time options to reduce energy-to-solution. This approach together with targetting optimal hw for a given problem would be aspects of a smart scheduler that determines where and how computational jobs would (re-compile &) run. Further, analysis of PUE of a given data centre and AI analysis of previous runs would enhance the scheduler to determine also when (and if!) a given job would run;
...the pedagogy of teaching energy efficient performant computing, bearing in mind not everybody has access to root level on a supercomputer, or even to a supercomputer. As part of this, I have set up MMU's "Greener Compute Club" where staff & students can engage in learning about, and help to reduce, the Carbon footprint of IT at the university, and how IT can be helped to reduce the Carbon footprint across the university and city.

Some aspects of these ideas are discussed in my recent CfACS seminar (PDF copy of slides) and form recent grant proposals to NWPST, NERC, UKRI. Further grants are planned to ESRC (sic) with support from Tyndall Centre, and EPSRC/NERC with University of Manchester.

My previous MSc/PhD students (inc. some at Liverpool) and their work, include

Konstantinos Panagiotis ALEXANDRIDIS - PhD candidate 2019-2023
Hamed ALGHAMDI - MSc 2019-2020 Dissertation, "The role of Cloud for Efficient HPC Pedagogy", Univ. of Liverpool (Oct 2020).
Matthew CARTER - MSc 2018-2019 Dissertation, "Determination of the Usability of FPGA Technology to Accelerate Option Pricing Algorithms", Univ. of Liverpool (Sept 2019). Matthew was a 2018-2019 winner of the "Ann Maybrey Prize" for excellent performance as evidenced by obtaining the highest overall marks during his MSc
Tsung-Ta "Jeffrey" HSIEH - MSc 2018-2019 Dissertation, "Determination of the Usability of FPGA Technology to Accelerate Process of Solving Eigenvalues and Eigenvectors", Univ. of Liverpool (Sept 2019)
Chris JONES - MSc 2019-2020/2020-2021, CFD on FPGA
Ryan LLEWELLYN - MSc 2019-2020 Dissertation, "Using ML for Improved Energy Efficient Computing", Univ. of Liverpool (Oct 2020).

My current & previous roles on research conferences include

Local Chair, "Parallel Numerical Methods and Applications" (Topic 9), Euro-Par 2022
Programme Committee member, Euro-Par 2023
Programme Committee member: Edu-Par 2022, NSF/TCPP Workshop on Parallel & Distributed Computing, held in conjunction with IPDPS 2022
Programme Committee member: EduHPC 2022, held in conjunction with SuperComputing 2022, EduHPC 2020 (@SuperComputing 2020)
Programme Committee member: EduPar 2024 (@IPDPS 2024), EduPar 2023, EduPar 2022, EduPar 2021 (@IPDPS 2021)
Chair, "Emerging Tech" conference 2019 (EMiT2019)
Programme Committee member: EMiT2016, EMiT2015, EMiT2014

I act as reviewer for a few supercomputing journals and for EPSRC.

My Teaching & "Office Hours" at Manchester Met

In the forthcoming academic year 2024-2025, I am Unit Lead for L6 and L7 High Performance Computing (HPC) & Big Data course units, and for the L6 Research in Computing, which these year has the theme "Green Software Engineering" (aka Greener Compute). I also am one of the tutors for Carbon Literacy optional of ICE (L6). Previously, I've written L5 Software Design & Architecture course which I wrote and delivered in 2022-2023. Indeed, in the academic year 2022-2023, I taught L5 Software Design & Architecture (as Unit Lead), L6 Programming Languages/Paradigms labs, and the MSc HPC & Big Data (as Unit Lead) course units.

If students wish to contact me, please just drop me an email me or send a message to me on MS Teams. You can do this at any time and I will reply within one of my working days. (Please note I am returning to part-time work shortly & will update this web page accordingly.) If you get an "out of office" then that will state my next working day or expected return to work.

My office hours during teaching weeks for semester 2 are 13:30-14:30 on a Tuesday and 09:30-10:30 on a Wednesday and Thursday. These exist to provide support and guidance. If you wish to meet then please let me me know in advance (so I can research any solutions if required) since I may be online rather than in the physical office. Alternative times (and times outside of teaching weeks) are available upon request.

Student Support Info

My History

My potted history includes:

lecturing in Computer Science at University of Liverpool
energy efficient computing research/R&D at STFC Hartree Centre
my own little company helping folks compute quicker and bigger with less energy at High End Compute
as part University of Manchester's Environmental Sustainability Team, authored review & plan on environmental impact of IT services

My Publications

2023: Martin Juckes, Michael Bane, Jennifer Bulpett, Katie Cartmell et al. Sustainability in Digital Research Infrastructure: UKRI Net Zero DRI Scoping Project final technical report. Zenodo. https://doi.org/10.5281/zenodo.8199984 (2023)
2023: Michael Bane, Oliver Brown, Teymoor Ali, Deepayan Bhowmik, Jamie Quinn, and David Stansby. ENERGETIC: Final Report v1.1. https://e-space.mmu.ac.uk/631671/ (March 2023).
2022: David Topping, Michael Bane, eds. Introduction to Aerosol Modelling: from theory to code. John Wiley And Sons Ltd. ISBN 978-1-119-62565-0
2022: Adi Kuntsman, Samuel Attwood, Michael K. Bane et al. Do digital technologies help or harm the planet: decolonising and decarbonising our digital society. Manchester Metropolitan University, U.K. (August, 2022). ISBN: 978-0-9933426-5-3. See RISE project report
2021: Benjamin C. B. Symons, Michael K. Bane, and Paul L. A. Popelier (2021). DL_FFLUX: A parallel, quantum chemical topology force field, Journal of Chemical Theory and Computation 2021 17 (11), 7043-7055. DOI: 10.1021/acs.jctc.1c00595
2020: J. C. Adams et al., "Lightning Talks of EduHPC 2020," 2020 IEEE/ACM Workshop on Education for High-Performance Computing (EduHPC), GA, USA, 2020, pp. 59-64, doi: 10.1109/EduHPC51895.2020.00013. which discusses SC20 presentation "Evolving the Traditional Student Cluster Competition as Tomorrow's "Peachy Assignments", EduHPC20: Workshop on Education for High-Performance Computing, Supercomputing 2020.
2019: "Proceedings of the 2019 Emerging Technology Conference". Editors: M.K.Bane and V.Holmes, ISBN 978-0-9933426-4-6
2019: UK OpenMP Users Conference: "Lessons from Coarse Grained OpenMP Parallelisation and Code Optimisations to Accelerate Quantum Computational Chemistry FORTRAN90 Code 'FFLUX'", Benjamin Symons and Paul Popelier, University of Manchester. Michael Bane, High End Compute Ltd
2018: Energy Efficient Computing Research at STFC Hartree Centre, Michael Bane, poster, ISC2018. https://2018.isc-program.com/presentation/?id=proj111&sess=sess144
2017: "Accelerating activity coefficient calculations using multicore platforms, and profiling the energy use resulting from such calculations", David Topping, Irfan Alibay, and Michael Bane, EGU2017-12246. (2017) https://meetingorganizer.copernicus.org/EGU2017/EGU2017-12246.pdf
2016: "FEREBUS: Highly Parallelized Engine for Kriging Training", N. Di Pasquale, M. Bane, S.J. Davie and P.L.A. Popelier, J. Comput.Chem., 37, 2606-2616 (2016).
2016: "Proceedings of the EMerging Technology (EMiT) Conference 2016. Editors: B.D.Rogers, D.Topping, F.Mantovani, M.K.Bane. ISBN 978-0-9933426-3-9"
2016: "UManSysProp: An online facility for molecular property prediction and atmospheric aerosol calculations", D.O. Topping, M. Barley, M. Bane, N.J. Higham, B. Aumont, and G. McFiggans, Geosci. Model. Dev. 9, 899-914, 2016
2015: Proceedings of the EMerging Technology (EMiT) Conference 2015. ISBN 978-0-9933426-0-8
2013: Case Study: BBC - In the Mood for Music"a case study of HTC (high throughput) work I undertook for BBC R&D to reduce time to analyse 128,000 music files from estimated 1.5 years to just 6 hours
2008: R. Warren, S. de la Nava Santos, N.W. Arnell, M. Bane, et al. "Development and illustrative outputs of the Community Integrated Assessment System (CIAS), a multi-institutional modular integrated assessment approach for modelling climate change", Environmental Modelling & Software, 23(5), 2008. (EMS "best paper" Award)
2006: R.W. Ford, G.D. Riley, M.K. Bane, C.W. Armstrong and T.L. Freeman (2006), "GCF: a General Coupling Framework", Concurrency and Computation: Practice & Experience, (John Wiley & Sons), vol. 18, no. 2, pp163-181. doi:10.1002/cpe.910. (PDF may require registration)
2005: Elliot, M. J., Manning, A., Mayes, K., Bane, M. and Gurd,J., (2005) "SUDA: a program for identifying and grading special uniques". Proceedings of UNECE worksession of statistical Data Confidentiality, Geneva, November 2005.
2002: Michael K Bane and Graham D Riley, "Extended Overhead Analysis for OpenMP" in Euro-Par 2002 Parallel Processing, 8th International EuroPar Conference, Paderborn, Germany, Aug 2002.
2001: M.K. Bane, M.D. Mihajlovic (2001), "A Fast Parallel Solver for the Biharmonic Problem", Proceedings of the Tenth SIAM Conference on Parallel Processing for Scientific Computing, Portsmouth, Virginia USA, 2001
2000: M.K. Bane and G.R. Riley, "Automatic Overheads Profiler for OpenMP Codes" EWOMP2000 conference, 14-15 Sept. 2000, University of Edinburgh.
2000: M.K. Bane, R. Keller, M. Pettipher & I. Smith (2000) "A Comparison of MPI and OpenMP Implementations of a Finite Element Analysis Code" Cray User Group 22-26 May 2000, Noordwijk NL
1992: M.K. Bane and T.L. Freeman, (1992) "Asynchronous Algorithms for Calculating Polynomial Zeros" in "Parallel Computing: Problems, Methods and Applications", ed. Messina, P. and Murli, A., Elsevier, Amsterdam, pp. 53-62.
1991: T.L. Freeman and M.K. Bane, "Asynchronous Polynomial Zero-Finding Algorithms". Parallel Computing 17, pp. 673-681, 1991.
1991: M.K. Bane and T.L. Freeman, "Implementation of Parallel Asynchronous Iterative Methods in occam" in "Applications of Transputers III", ed. Durrani, T.S., et al, I.O.S., Amsterdam, 1991.
1990: T.L. Freeman and M.K. Bane, "An occam Implementation of an Asynchronous Algorithm for Calculating Polynomial Zeros" in "Applications of Transputers II", ed. Pritchard, D.J. and Scott, C.J., I.O.S., Amsterdam, pp. 533-540, 1990

My Contact Details

email // LinkedIn // Twitter // MS Teams

MICHAEL BANE [HOME] [Green Software Engineering] [Greener Compute Club]