The aim of this testbed is to ensure that:
- ARM servers work harmoniously with accelerators (such as GPUs)
- Any shortcomings are understood, documented and reported to vendors
- ExCALIBUR and the wider UK research community has access to an ARM-GPU testbed
The ARM+GPU testbed system is based on HPE’s Apollo 70 platform and Marvell’s ThunderX2 processors, with 2x NVidia V100 GPU per server, incorporated into the University’s existing ARM Catalyst system.
The project includes Research Software Engineering (RSE) effort for the porting, benchmarking and development of existing codes to support the programme of work described above. The RSE will also contribute to the creation of digital assets including progress reports, whitepapers and how-to documents, as well as software enhancements and modification. The ARM+GPU testbed builds on the University of Leicester’s existing experience managing ARM based HPC systems, our Software engineering team’s CUDA expertise, and our existing relationship with the technical teams at ARM.
Access to the testbed systems
The Excalibur H&ES testbeds are prioritised for access by ExCALIBUR projects, but also available for use by the wider UK research community - contact the ExCALIBUR H&ES programme office to discuss your requirements. Please note that the testbeds are offered on a best efforts basis rather than a service footing, as befits their experimental status.