One of the conclusions of Driving UK HPC enabled science was that there was a clear need for wider engagement with the US HPC community rather than being limited to a small number of sites or one geographical region. Therefore, last year @UKUSAHPC proposed to coordinate UK engagement at the 2013 Supercomputing Conference (SC13), where US national laboratories, academia and industry are all well represented. As part of this initiative @UKUSAHPC awarded 10 travel grants predominantly to early career researchers to support the pump-priming of UK/USA collaborations by facilitating one-to-one interactions and increasing awareness of research opportunities with the US community.
As expected, attending SC13 proved to be a highly effective strategy to grow the professional networks between UK and USA academia, government and industry. All of the UK delegates reported multiple new interactions with excellent potential for collaboration, as well as having developed a good knowledge of the funding and HPC resource landscape in the US. This community engagement complemented existing UK persons present such as those representing the UK Regional HPC Centres, Hartree Centre, ARCHER and HPC-SIG.
Access to supercomputers is an important issue for anyone involved in research that demands HPC. US collaborators not only contribute their knowledge and skills, but they can provide a route to world leading supercomputing resources such as those in XSEDE. While the focus in HPC often tends to be on big iron and the Linpack benchmark, it is actually the individual scientists working together that create the impact the hardware was created for. This further underscores the importance of supporting individual collaborations at the grass roots level.
From both the UK and US communities a common message during SC13 was confusion over who best to contact for exploring collaborative opportunities. High-level interactions between RCUK and US counterparts such as the NSF and DOE are vital to give strategic oversight and get appropriate agreements in place, and indeed these discussions are ongoing. However, there are also other independent bodies such as CASC (a coalition of 79 US academic institutes for scientific computation) who also play a vital role in the US HPC community. CASC kindly invited @UKUSAHPC and UK HPC-SIG (which is the nearest comparable organisation in the UK) to their Luncheon at SC13 to explore how cooperation and communication might be improved between the UK and US HPC communities. As it was apparent that there was a great deal of interest on both sides to follow up on discussions, @UKUSAHPC also supported a HPC-SIG delegation to attend the CASC Spring meeting with the objectives:
So what is next for @UKUSAHPC? We are currently in the phase of attempting to secure a successive year of funding. We want to extend our support of early career researchers in HPC. For example:
As expected, attending SC13 proved to be a highly effective strategy to grow the professional networks between UK and USA academia, government and industry. All of the UK delegates reported multiple new interactions with excellent potential for collaboration, as well as having developed a good knowledge of the funding and HPC resource landscape in the US. This community engagement complemented existing UK persons present such as those representing the UK Regional HPC Centres, Hartree Centre, ARCHER and HPC-SIG.
Access to supercomputers is an important issue for anyone involved in research that demands HPC. US collaborators not only contribute their knowledge and skills, but they can provide a route to world leading supercomputing resources such as those in XSEDE. While the focus in HPC often tends to be on big iron and the Linpack benchmark, it is actually the individual scientists working together that create the impact the hardware was created for. This further underscores the importance of supporting individual collaborations at the grass roots level.
From both the UK and US communities a common message during SC13 was confusion over who best to contact for exploring collaborative opportunities. High-level interactions between RCUK and US counterparts such as the NSF and DOE are vital to give strategic oversight and get appropriate agreements in place, and indeed these discussions are ongoing. However, there are also other independent bodies such as CASC (a coalition of 79 US academic institutes for scientific computation) who also play a vital role in the US HPC community. CASC kindly invited @UKUSAHPC and UK HPC-SIG (which is the nearest comparable organisation in the UK) to their Luncheon at SC13 to explore how cooperation and communication might be improved between the UK and US HPC communities. As it was apparent that there was a great deal of interest on both sides to follow up on discussions, @UKUSAHPC also supported a HPC-SIG delegation to attend the CASC Spring meeting with the objectives:
- Broadening communication between CASC and UK HPC-SIG for the purpose of fostering cooperation and collaboration.
- Increase awareness within UK HPC-SIG of opportunities in US HPC: as UK HPC-SIG members are closely connected with the UK HPC user community they provide excellent dissemination points for this information.
- UK HPC-SIG to investigate if there are lessons to be learned from CASC that could increase the impact of HPC in the UK. For example, CASC is highly effective at communicating the impact of computational science and HPC to industry and government.
So what is next for @UKUSAHPC? We are currently in the phase of attempting to secure a successive year of funding. We want to extend our support of early career researchers in HPC. For example:
- Rerunning the travel award scheme for SC14.
- Open call for UK researchers to visit US national labs and universities.
- Open call to invite US researchers to the UK on research visits.
This list is not exhaustive, so if you have a good idea please or would like to be involved then please email us or tweet @UKUSAHPC.
