Intel has rolled out the new Knights Landing Xeon Phi supercomputing processor, which it claims will deliver more than 3 TFLOPS of double-precision performance and three times the single-threaded performance than the current generation.
Scheduled to power high-performance computing (HPC) systems during the second half of 2015, the new chip will be offered as a standalone processor integrated directly on the motherboard socket in addition to the PCIe-based card option.
Powered by about 60 HPC-enhanced Silvermont architecture-based cores, Knights Landing will support DDR4 system memory equivalent in capacity and bandwidth to Intel Xeon processor-based platforms.
As part of efforts to re-architect the fundamental building block of HPC systems, the chipmaker has launched the new Intel Omni Scale Fabric for integration into Knights Landing, which could jointly offer rapid data relevant products/services transfers.
Intel Workstations and HPC VP and GM Charles Wuischpard said that the company is re-architecting the fundamental building block of HPC systems by integrating the Intel Omni Scale Fabric into Knights Landing, marking a significant inflection and milestone for the HPC industry.
"Knights Landing will be the first true many-core processor to address today’s memory and I/O performance challenges," Wuischpard said.
"It will allow programmers to leverage existing code and standard programming models to achieve significant performance gains on a wide set of applications.
"Its platform design, programming model and balanced performance makes it the first viable step towards exascale."
Built upon a combination of improved acquired IP from Cray and QLogic, the new Omni Scale fabric integrates a complete product range comprising adapters, edge switches, director switch systems, and open-source fabric management and software tools.
In addition to being integrated into the Knights Landing processor, the new Omni Scale fabric will also be integrated into the chip maker’s ‘future 14nm Intel Xeon processors’.
