Competition in the HPC processor market is building with the news that Adapteva, a specialist in the design of energy efficient accelerator chips and compute modules, has created a 1024 core processor – thanks, in part, due to a grant from the US, Defense Advanced Research Projects Agency (DARPA).
This new processor is aimed at emerging workloads such as deep learning, self-driving cars, autonomous drones, which ‘require an order of magnitude boost in processing efficiency to unlock their true potentials’ according to Andreas Olofsson, CEO, and Founder at Adapteva.
The chip ‘Epiphany-V’ contains an array of 1024 64-bit RISC processors, 64MB of on-chip SRAM, three 136-bit wide mesh Networks-On-Chip, and 1024 programmable IO pins. The chip has taped out and is being manufactured by TSMC.
Olofsson explained that the primary goal of the project was ‘build a parallel processor with 1024 RISC cores demonstrating a processing energy efficiency of 75 GFLOPS/Watt. A secondary goal of this project is to demonstrate a 100x reduction in chip design costs for advanced node ASICs.’
Adapteva released a paper detailing the technology which explains that the Epiphany architecture is a ‘distributed shared memory architecture comprised of an array of RISC processors communicating via a low-latency mesh Network-on-Chip.’
Each node in the processor array is a complete RISC processor capable of running an operating system (“MIMD”). Epiphany uses a flat cache-less memory model, in which all distributed memory is readable and writable by all processors in the system.
This is the fifth generation of the Epiphany product line and includes new capabilities compared to previous iterations including; 64-bit memory addressing, 64-bit floating point operations, 2X the memory per processor, and custom ISAs for deep learning, communication, and cryptography.
Now that the research has been completed Olofsson commented that the next stage of the project would be to 'fully characterise the Epiphany-V silicon devices once devices return from the foundry.'
'Future work will focus on extending and customising the Epiphany-V SOC platform for specific target applications’ concluded Olofsson.
