While supercomputers are moving towards Exascale and steadily becoming more powerful every year, they are also becoming more difficult to program. EPiGRAM-HS is a European Commission Funded project (Horizon 2020, Grant agreement ID: 801039) with the goal of designing and delivering a programming environment for Exascale heterogeneous systems in order to support the execution of large scale applications.
Exascale supercomputers are capable of an exaFLOP (a billion billion floating point operations per second) and is the next milestone in computing power of large scale computing systems. In order to burst their compute power, these machines are using a property known as heterogeneity, meaning that there are more than one kind of processor or cores like GPU’s and FPGA’s
The underline complexity of heterogeneous hardware creates drawbacks for application developers who are often not aware of the hardware specifics and therefore cannot adjust their code to take advantage of the accelerating devices in heterogeneous systems.
EPiGRAM-HS consists of 6 partner organisations that are working towards the delivery of a programming framework that will optimise the work of the application experts and give their applications the opportunity to use the extreme compute power of an Exascale machine.
Network. Exploiting heterogeneity for high performance communication,, building on proven programming models.
Compute. Simplified efficient usage of complex memory systems.
Memory. Improving the integration of code for leading edge accelerators in work flows in HPC applications.
EPiGRAM-HS has a set of 4 pilot applications that will set the requirements by describing extreme-scale application bottlenecks and demonstrating improvements to application performance by targeting the EPiGRAM-HS environment.
More productive and powerful software has a strong societal impact on many levels. Weather forecast can be faster and more precise and therefore potentially save lives, as well as large amounts of public funds ahead of climate change. Space weather and computational fluid dynamics can get important data and help create more effective technologies for the future. Finally, lung cancer detection using deep learning applications will be of profound importance in the future for public hearth health care and the transition to e-health.