Intel’s next-generation HPC fabric, based upon the Intel® Omni-Path Architecture (Intel® OPA) will be officially announced later this quarter. It will be a critical ingredient in the Intel scalable system framework, a flexible blueprint for developing high performance, balanced, power-efficient and reliable systems capable of supporting both compute- and data-intensive workloads. “To create the next generation of highly efficient supercomputers, we need to make sure the right ingredients are combined in precisely the right way,” stated Al Gara, Intel Fellow and Chief Exascale Architect for Intel’s Technical Computing Group. “To do that, we must look at things from a holistic view, from a total system perspective. And out of that comes the definition and the development of those ingredients.”
Intel OPA builds on enhanced versions of technologies, like Intel® True-Scale Fabric InfiniBand, Cray Aries and the OpenFabric Alliance host stack. To this foundation Intel added revolutionary new features designed to enhance the fabric’s switching environment. The result is improved QoS, reliability, performance and scalability. The fabric is designed to support the next generations Intel processors, both Intel Xeon® E5 family of processors and the many-core, highly parallel processors such as Knights Landing and Knights Hill, both members of the Intel® Xeon Phi™ family.
Intel® OPA 48 port Edge Switch Intel® OPA standard PCI board
These devices have not been authorized as required by the rules of the Federal Communications Commission, including all Intel® Omni-Path Architecture devices. These devices are not, and may not be, offered for sale or lease, or sold or leased, until authorization is obtained.
Intel is building upon the work of the Open Fabric Alliance* (OFA) and the OFA Enterprise Distribution by enhancing the OFA distro with Omni-Path technology including device drivers and key software for management and support. Compatibility and stability are the result. Intel OPA is a mature software stack that provides users with compatibility between their current technology and the Omni-Path architecture.
Another essential strategy is to drive the fabric closer to the CPU until it becomes essentially an extension of the CPU itself. This integration improves over time. The initial Intel OPA release will utilize discrete adapters, but quickly it will become integrated into Intel Xeon Phi and then Xeon processors that will allow Intel to increase the bandwidth to each Intel Xeon Phi socket. Intel OPA delivers up to 50GB of bi-directional bandwidth in order to feed hungry processors and remove performance bottlenecks.
To educate the HPC community the features and benefits of this upcoming product, Intel has developed a series of webinars focusing on the following topics:
- Part 1 – Intel’s Host Fabric Architecture: Architected for Your MPI Applications
- Part 2 – Transitioning to the Intel® Omni-Path Architecture
- Part 3 – Designing Fabrics with Intel Omni-Path Architecture
- Part 4 – Next-generation fabric: New public details on the Intel Omni-Path Architecture
- Part 5 – Intel Omni-Path Architecture Storage Strategy
- Part 6 – Advanced Features of the Intel® OPA Network Layers
Click here to view any of the videos in this series. By registering at this site, you will also receive an invitation to a special November webinar featuring guest speaker Steve Conway of IDC Research.