Cray cracks heterogeneous supercomputing

Cray cracks heterogeneous supercomputing


New machines will integrate a range of processing technologies in a single platform

Supercomputer manufacturer Cray has announced plans to develop machines that take the concept of heterogeneous computing to a new level by integrating a range of processing technologies in a single platform.

'Adaptive supercomputing' systems, to be rolled out over the next few years, will combine standard microprocessors using scalar processing, with vector processing, multithreading and hardware accelerators in one high-performance computing (HPC) platform running Linux.

Compilers and other software will automatically match an application to the processor blade best suited to the task.

As a result, users will not have to bias their HPC platform decisions toward the most widely used applications, and will not have to make programming alterations to allow their applications to run efficiently on a particular platform.

"Different applications run best on different types of processors, but high-performance computers typically offer only one type of processor," said Cray chief technology officer Steve Scott.

"Even today's heterogeneous computing environments just loosely link differently architected computers, rather than offering true processing heterogeneity and adaptability.

"Cray will build supercomputers that can adapt to the applications, instead of forcing the applications to adapt to the supercomputers."

Scott added that, over time, the systems will include intelligence that can examine an application, determine which processing technique will work best and then handle the application accordingly, all without user intervention.

Cray will implement adaptive supercomputing in phases. The first phase, codenamed Rainier, will create an integrated user environment across all of Cray's platforms.

The second phase will result in a fully integrated multi-architecture system, and the final phase will develop systems that incorporate dynamic resource allocation using software that automates adaptive supercomputing.