SeRC Turns to oneAPI Multi-Chip Programming Model for Accelerated Research

Print Friendly, PDF & Email

At ISC 2020 Digital, the Swedish e-Science Research Center (SeRC), Stockholm, has announced plans to use Intel’s oneAPI unified programming language by researchers conducting  massive simulations powered by CPUs and GPUs.

The center said it chose the oneAPI programming model, designed to span CPUs, GPUs, FPGAs and other architectures and silicon,  to accelerate compute for research using GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics software, developed by SeRC and first released in 1991.

SeRC said it will utilize CPUs and GPUs from Intel, Nvidia and AMD to run HPC simulations of large biomolecules and to study how proteins and lipids function on the atomic level, such as when binding other molecules or to design new drugs targeting COVID-19.

“Simulation coupled with exascale compute allows scientists unprecedented visibility into the molecular mechanisms of life and provides a computational microscope that enables them to investigate what-if scenarios that would be difficult if not impossible to test in a traditional lab,” SeRC said. “Many biological processes studied by molecular dynamics are in the microsecond timescale. MD time-steps are around a femtosecond and therefore ~10^9 simulation steps are required. To finish a simulation in about a week requires each simulation step to be computed in under a millisecond. This is only possible with very efficient communication between CPUs and GPUs working together.”

Researchers at Stockholms universitet (SU) and Kungliga Tekniska högskolan (KTH), two of the four universities that form SeRC, will work with academic groups and also with Intel engineers on oneAPI software optimizations.

Professor Erik Lindahl

“The future of hardware is heterogeneous,” said Erik Lindahl, professor of biophysics at KTH & SU, “which makes it critical to move towards open performance-portable APIs, and we are particularly happy with how the ecosystem from the start has been closely involved in the development and steering of oneAPI.”

“It is exceptionally important to us that GROMACS is able to make efficient use of all the fastest supercomputers in the world,” Lindahl said, “and the upcoming exascale machines powered by oneAPI will make it possible to simulate processes we could not even imagine a few years ago, such as how a virus binds to proteins on a cell and infects it.”

With much fanfare and to some skepticism, Intel announced oneAPI in December 2018. Industry analyst Patrick Moorhead, president of Moor Insights & Strategy,  called it Intel’s “most ambitious software project.”

“I like to call oneAPI the ‘magic API,’ given its potential,” Moorhead said in an article in Forbes magazine last year. “oneAPI…are tools and libraries that abstract the CPU, GPU, FPGA, and AI accelerators.  So instead of writing to four or five different API’s, you are writing to one. Today, if you are a company which programs for CPU, GPU, FPGA, or ML accelerator, you are using different toolsets across every single accelerator… (oneAPI) will include a unified language based on Data-Parallel C++ along with libraries to deliver native code performance. This will allow for greater optimization of middleware and frameworks in areas like AI…. oneAPI will be very difficult to create, but if Intel can pull it off, it will be very valuable to developers, researchers, and businesses alike….”