High Performance Edge Computing - The Frontier for Supercomputing and Acceleration

What is High Performance Edge Computing?

High Performance Edge Computing (HPEC) means bringing high-performance computing capabilities - typically found in data center infrastructures - into hardware systems deployable at the edge, where environmental conditions are typically harsh. This requires a rugged and compact design to ensure reliable and sustained operations in a wide range of temperatures and in difficult shock and vibration conditions.

Edge computing - and, when required, High Performance Edge Computing - is the new domain for innovation. Next-generation analytics powered by artificial intelligence (AI), machine learning (ML), and other high-performance workload processing require comprehensive intelligent edge frameworks and platforms. Furthermore, these platforms must have maintained pace with the convergence of embedded/OT and enterprise/IT domains through the past several years.

High Performance Edge Computers feature distinct development requirements and challenges spanning hardware, software, connectivity, platform integrations, and security. Years ago, a few suppliers had the foresight and ambition to begin tackling HPEC and forming a partnership ecosystem including key software and technology providers. Those that did, however, are now reaping the fruits of their labor and have a dramatic head start over competitors scrambling to take a stake in a critical piece of the market for many different industries.

What are the challenges of High Performance Edge Computing architectures?

A major challenge still facing embedded technology suppliers and manufacturers is the end-to-end integration of connected systems, cloud endpoints, and third-party platforms or services. A flexible embedded framework is required to ensure maximum use of data generated/collected over the long term, particularly for fueling business applications and cross-integration with different IoT platforms or PaaS/SaaS (Platform-as-a-Service or Software-as-a-Service).

This is a necessary precursor to developing Edge servers and processing solutions. The framework must thread through the OT and IT domains to include support for various field infrastructure, communications infrastructure, (IoT) application infrastructure, as well as the application/enterprise IT layer. Without a flexible edge-to-cloud integration platform and supporting software/middleware libraries for IoT edge gateways and other connected systems, solutions for high-performance edge computing cannot scale or adapt to the dynamic requirements of solution providers and end users.

High performance edge infrastructure is drastically changing edge-to-cloud solution architectures in the development of new use cases leveraging connected data streams. New classes of hardware, software, and IP are needed for hosting increasingly popular accelerated processing, machine learning/inference, and vision applications at the edge. In some instances, these growing requirements are fueling the demand for more vertically tailored features, functionality, computing profiles, and form factors. The intelligent edge requires a healthy integration of OT and IT infrastructure to stay at pace with system and solution requirements.