One of the challenges for high-performance computing is to enable organizations to access massive amounts of data and deliver that to applications regardless of where these sit in the enterprise. This is a complicated problem to solve in the current hybrid computing world, and Sycomp, A Technology Company Inc., a global provider of data center, cloud and security services, has been working with IBM Corp. and Google LLC to find a solution.

One outcome was released on Google Cloud Marketplace in March. It is Sycomp Storage fueled by IBM Storage Scale to provide a performant parallel file system that can dynamically access data in Google Cloud Storage and deliver it with up to 320 gigabits of low latency read capability. The solution allows users to simultaneously access data from hundreds of virtual machines.

“With IBM Cloud and Google Cloud … they’ve built the networks, they’ve built the infrastructure,” said John Zawistowski (pictured, left), global systems solutions executive at Sycomp. “The challenge is data … how do we get it to the application regardless of where it is? That’s what we’ve worked on with Google and IBM Cloud to bring them a solution that allows that data highway between anywhere and anyplace.”

Zawistowski spoke with theCUBE industry analyst John Furrier at SC23, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. He was joined by Hugo Saleh (right), head of HPC at Google, and Leon Platts (center), vice president of IBM Cloud high-performance computing at IBM, and they discussed the challenges in HPC and how the three companies collaborated on an innovative solution. (* Disclosure below.)

Playing Tetris

The collaborative work between Sycomp, Google and IBM comes at a time when grid managers must find innovative ways to link disparate systems into large infrastructures, a process that IBM’s Platts compares to Tetris, a popular puzzle video games created in the 1980s.

“One of the challenges that I see with clients day-in and day-out is they’re playing Tetris with their grids,” Platts said. “They’ve got 20 different product managers that need access to that grid to deliver on a product. All of these grids are doubling, tripling, quadrupling by 2030. My heart goes out to those grid managers … how we can help them at IBM, how we can work with our ecosystem, with Google and Sycomp, and how we can deliver to them the performance and ability to play better Tetris.”

Playing better Tetris also means adopting a solution that can meet the needs of businesses regardless of size. Sycomp’s jointly developed solution is tailored to meet the throughput needs for a wide range of use cases, according to Zawistowski.

“We’re delivering for hedge fund clients,” he said. “They wanted … 300 gigs per second throughput. Well, we delivered that, but then you get the other end of the spectrum where I only need 25 gigs or 10 gigs. So, our solution can span that gap.”

For HPC practitioners, the challenges of throughput, latency and data movement come with the territory, and expanding interest in AI will drive new solutions in the years ahead.

“Challenges and constraint hurdles are Monday morning in HPC,” Saleh said. “We’ve been tearing those down for decades now. The demand for HPC, the demand for AI is just straining every aspect of the ecosystem to build more data centers, to build more power plants, make it green. Those are challenges that as an industry and an ecosystem we’re going to have to tackle together.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of SC23:

(* Disclosure: Sycomp, A Technology Company Inc. sponsored this segment of theCUBE. Neither Sycomp nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE