Frontier, or OLCF-5 was build by Hewlett Packard Enterprise for the US Department of Energy's Oak Ridge National Laboratory and may tackle problems that are eight times complex up to 10 times faster when put next to other supercomputers.
A typical laptop is barely capable of some teraflops, or a trillion operations per second, which could be a million times less. The exaflop machine, called Frontier, could help solve a variety of complex scientific problems, like accurate climate modelling, nuclear reaction simulation and drug discovery.
“Frontier will offer modelling and simulation capabilities at the very best level of computing performance,” says Thomas Zacharia at Oak Ridge National Laboratory.
The Frontier computer system is housed in 74 separate cabinets, comprising 9400 CPUs, or standard computer processors, and 37,000 GPUs, which are processors designed to render 3D graphics but can even be used for a variety of other tasks.
This means the machine features a total of 8,730,112 cores capable of parallel computing tasks; a typical laptop has between five and nine. At peak power, the pc generates most heat that it requires four high-powered pumps to push quite 25,000 litres of water round the machine each minutes.
Frontier’s exaflop performance means not only is it favorite within the TOP500, a world collaboration to rank the world’s most powerful supercomputers, but it also represents 1 / 4 of the computing power of the whole list.
“One machine represents 25 per cent of the overall performance of the entire list, so it’s a really, very impressive achievement,” says Simon McIntosh-Smith at the University of Bristol, UK.
Frontier has yet to realize its final form, either. Over the approaching months and years, as its software is optimised, it could reach a theoretical peak of two exaflops.
In the past, supercomputing milestones are shortly followed by more machines with similar capabilities. While there are several exascale machines planned for the following number of years, it isn’t clear how widespread this technology might become.
“The rate of improvement in electronics has bogged down somewhat, so we don’t expect exascale machines to proliferate through the TOP500 as quickly because it did for petascale, as an example,” says McIntosh-Smith, concerning machines with a thousandth of the capabilities.
Frontier has coherent interconnects between CPUs and GPUs, allowing GPU memory to be accessed coherently by code running on the Epyc CPUs.
Frontier also performed exceptionally well on computing benchmarks. This category is officially called "mixed-precision computing" and rates performance in formats that are generally used for computer science, wherein the pc showed accelerates to six.88 exaflops. "At an exascale speed, Frontier’s users can develop AI models that are 4.5X faster and 8X larger, allowing to coach more data which will increase predictability and speed time-to-discovery," the presser stated.
Frontier is constructed with HPE Cray EX supercomputers that deliver end-to-end capabilities comprised of compute, accelerated compute, software, storage and networking to support the magnitude of exascale performance.
What are Frontier's applications?
According to its creators, the Frontier supercomputer would have significant impact in areas of importance like cancer and disease diagnosis. The list goes on! Frontier could help with prognosis, drug discovery, renewable energy, and new materials to develop safer and sustainable products.
When researchers gain access to the fully operational Frontier system later this year, it'll mark the culmination of labor that began over three years ago involving many talented people across the Department of Energy and our industry partners at HPE and AMD,” said Jeff Nichols, associate lab director, Computing and Computational Sciences, Oak Ridge National Laboratory. “Scientists and engineers from round the world will put these extraordinary computing speeds to figure to unravel a number of the foremost challenging questions of our era, and plenty of will begin their exploration on Day One.
You must be logged in to post a comment.