High Performance Computing (HPC)

Project Lead: Adi Gangidi (Rackspace), Steven Roberts (IBM)

IC Representative: Phil Pokorny (Penguin) 

The HPC Project is focused on developing a fully open heterogeneous computing, networking and fabric platform Optimized for a multi-node processor that is agnostic to any computing group using x86, ARM, Open Power, FPGA, ASICs, DSP, and GPU silicon on an hardware platform.  This enables rapid innovation in low latency high Performance Computing and Big Data analytics through open non-lock-in computing, interconnect, and software stack.

HPC Project Scope:

• Low latency top of rack switching

• Combined Switch and micro servers

• Combined compute and switching

• Low latency scalable storage

• Connectivity from HPC Fabrics/clustering technology to out of cluster networking via ethernet

• OCP mechanicals (19 and 21 inch)

• APIs and software interfaces

• Open Hardware compute and switching Development of OCP HPC Interconnect silicon spec (can build on existing mainstream technologies such as PCIe, RapidIO, Infiniband, Ethernet)