Oct. 18 is Exascale Day, and  Hewlett Packard Enterprise chose the date for a reason: Exascale is the point at which high-performance computing makes the leap to 10¹⁸ calculations per second. Although there are not yet any exascale computers in operation, the technology is close.

The Frontier computer, a joint United States Department of Energy and Hewlett Packard Enterprise Co. subsidiary Cray Inc. project, is scheduled to be performing calculations at speeds in excess of 1.5 exaflops before the end of 2021.

“There are areas in terms of weather prediction, climate prediction, drug discovery, material knowledge [and] engineering problems that are going to be unlocked with the use of exascale class systems,” said Ben Bennett (pictured), director of HPC strategic marketing at HPE.

Bennett spoke with Dave Vellante, host of theCUBE, SiliconANGLE Media’s livestreaming studio, during Exascale Day 2020. They discussed the history of HPC, the benefits that exascale computing will bring and the problems still to overcome. (* Disclosure below.)

Faster computational speeds provide more insights

The purpose of computing is not data; it’s insights, according to Bennett. And exascale computing promises more insights, faster. One highly anticipated use is the ability to streamline the drug production pipeline through computer model testing instead of lengthy and expensive human trials.

“Imagine having a genetic map of very large numbers of people on the Earth and being able to test your drug against that breadth of person, and knowing that 99% of the time it works fine,” Bennett said. “If you’re confident in your testing, if you can demonstrate that you can keep the one percent away for whom that drug doesn’t work, bingo! You now have a drug for the majority of the people.”

Current Federal Drug and Food Administration regulations prohibit the sale of drugs without human tests. But while the HPC world has been working hard to find a cure for COVID, exascale computing still has some hurdles to overcome before it can join the race.

“Are we at a point where we still need human trials? Yes. Do we still need due diligence? Yes,” Bennett stated. “Exascale is coming, [but] it’s not there yet.”

Virtual explosions remove the need to destroy physical prototypes

Another area that will benefit from exascale computing is mechanical safety testing. For example, all new jet engine designs have to demonstrate that the cowling is strong enough to withstand impact from a fan blade. Current petascale computational speeds aren’t fast enough to accurately model this scenario, according to Bennett. So, the manufacturer has to build a test engine at a cost of tens of millions of dollars and then try to destroy it.

“They run the engine at full speed, and then they literally fire off a little explosive and [blow] a fan blade off to make sure that it doesn’t go through the cowling,” Bennett said.

Computational modeling on exascale computers will eliminate the need to build and test a physical prototype, reducing the expense and time spent on testing a product.

While exascale is the current pinnacle that HPC is scaling, it isn’t the end of the journey, Bennett predicted.

“Ultimately it’s our quest for knowledge that drives these machines,” he said. “In 10, 20, 30 years time, I should be able to look back … and look at much faster machines and say: ‘Do you remember the days when we thought exascale was fast?’”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of Exascale Day 2020. (* Disclosure: TheCUBE is a paid media partner for Exascale Day 2020. Neither Hewlett Packard Enterprise, the sponsor for theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE