Chip-maker Intel has begun shipping of its new Xeon Phi processors, showcased last year, targeted at High Performance Computing (HPC) applications.

"These are high energy efficient performance processors; they pack a lot of performance per unit of energy used and they are very easy to programme," Rajeeb Hazra, Vice-President, data centre group, and General Manager, enterprise and government group at Intel, said.

"There are a lot of applications these processors can be put to use, including traffic management and weather prediction, while the greatest place these will find use is in machine learning," he added.

The company has begun the shipping of the processor, which was launched under its Knights Landing series at the ISC High Performance Computing 2016 in Frankfurt, Germany.

These can be used as standalone processors or co-processors, typically for an Intel Xeon, and can be used in supercomputers and servers.

As data volumes continue to explode and become more complex, new hardware, software and architectures are needed to drive deeper insight and accelerate new discoveries, business innovation and the next evolution of analytics in machine learning and the field of artificial intelligence, Intel said in a statement,

"A key to unlocking these deeper insights is the new Intel Xeon Phi processor," it added.

The Xeon Phi processor is Intel’s first bootable host processor specifically designed for highly parallel workloads, and the first to integrate both memory and fabric technologies. The chip is a general purpose central processing unit (CPU) built on open standards, making software investments portable into the future, it added.

Intel also unveiled HPC Orchestrator, a pre-integrated software stack that can power HPC clusters, which will be available from the fourth quarter of 2016.

Xeon Phi and HPC Orchestrator are parts of Intel's Scalable System Framework (SSF) architecture for HPC.

(The correspondent is in Frankfurt at the invitation of Intel.)

rajesh.kurup@thehindu.co.in

comment COMMENT NOW