Documentation

Across a wide range of industries, data acceleration is the key to building efficient, smart systems. A number of accelerator technologies have appeared to fill the gap that are based on custom silicon, graphics processors or dynamically reconfigurable hardware, but the key to their success is their ability to integrate into an environment where high throughput, low latency and ease of development are paramount requirements. A board-level platform developed jointly by Achronix and BittWare has been optimized for these applications, providing developers with a rapid path to deployment for high-throughput data acceleration.

Devices aimed at addressing modern algorithm acceleration workloads must be able to efficiently move high-bandwidth data streams between high-speed interfaces and throughout the device. Achronix Speedster®7t FPGAs can process these high-bandwidth data streams via an integrated new and highly innovative two-dimensional network on chip (2D NoC). This white paper discusses two methods of implementing a 2D NoC and presents an example design to show how the Achronix 2D NoC improves performance, reduces area, and reduces design time when compared to a soft 2D NoC implementation.

Deep learning's demand for computing power is growing at an incredible rate, accelerating recently from doubling every year to doubling every three months. Increasing the capacity of deep neural network (DNN) models has shown improvements across a wide range of areas ranging from natural language processing to image processing. This growth calls for the adoption of customized architectures that squeeze the greatest amount of performance out of each transistor available.

AI requires a careful balance of datapath performance, memory latency, and throughput that requires an approach based on pulling as much of the functionality as possible into an ASIC or SoC. But that single-chip device needs plasticity to be able to handle the changes in structure that are inevitable in machine-learning projects. Adding eFPGA technology provides the mixture of flexibility and support for custom logic that the market requires. Achronix provides not only the building blocks required for an AI-ready eFPGA solution, but also delivers a framework that supports design through to debug and test of the final application. Only Achronix Speedcore IP has the right mix of features for advanced AI that will support a new generation of real-time, self-learning systems.

With the technology industry at a crossroads — the effective repeal of Moore's Law  — data centers have become the sweet spot of the technology sector, showing healthy revenue growth and attracting new system solutions in both hardware and software. Unlike the ethereal promise of upcoming wonders from AI, robotics and the IoT, data center growth and innovation is happening in the here and now, with an even brighter future ahead the moment other nascent markets emerge from their chrysalis with killer apps of their own.