Webinars

Unlocking the Full Potential of FPGAs for Real-Time Machine Learning Inference

Machine Learning (ML) inference requires a performant overlay to transform the FPGA from a generic solution into a highly capable AI inference accelerator. In this presentation, using the example of automatic speech recognition (ASR), our Sr. Manager of Product Planning, Salvador Alvarez, explores how an overlay can be used to optimally leverage the potential performance of an FPGA architecture.

The Rise of the SmartNIC

Almost all of our information is now being digitized, forcing us to reimagine how data centers are built, down to the hardware and systems levels. An emerging data center trend is the rise of the SmartNIC: a programmable accelerator that accelerates and offloads an increasing number of tasks from the server CPU.

eFPGA IP Workload for Autonomous Vehicles

To unlock Level 5, truly autonomous vehicles, the industry needs more than just eccentric machine learning algorithms. The future of the automotive industry is dependent on our ability to produce high-performance, low-latency computing platforms.

FPGA Technology Enables Industry 4.0 and 5.0 through AI/ML Acceleration

Throughout history, the factory has been at the center of the world’s technological innovation. From the times of the early Industrial Revolution to today, manufacturing has seen significant advancement, and this trend is far from over.

Managing FPGA Resources as Virtualized Accelerator Blocks

With the convergence of processing requirements across multiple different industry verticals, the need to support flexibility and scalability in new designs is even more essential. Whilst virtualized/containerized workloads running on CPUs are central to, and simplify many of these new designs, the need for workload acceleration is essential to meet overall power, throughput and latency requirements.

How FPGAs Enable Heterogeneous Computing

Today there is a wide spectrum of computing models that address areas that have moved beyond the realm of programmatic CPU-based workloads. This area is called heterogeneous computing. It spans the realms of data centers to telco edge to embedded devices. The heterogeneous computing model encompasses CPUs, GPUs, and hardware accelerators. Field Programmable Gate Arrays (FPGAs) provide a technology in which flexible hardware accelerators can be realized.

How to Overcome the Pain Points of AI/ML Hardware

Enhance IoT and 5G Experience Through FPGA Data Acceleration

Worldwide data is forecasted to grow exponentially to 175ZB over the next 4 years with over 30 billion IoT connected devices being a significant contributor to data growth. 5G technology enables this growth by allowing data to move through the network faster and with substantially less latency.

Five Reasons Why a High Performance Reconfigurable SmartNIC Demands a 2D NoC Webinar

The current generation of SmartNICs relies heavily upon utilizing semiconductor devices with many ARM cores (or similar) to process packets at 25GbE. This approach, which is already challenged at 25GbE, becomes even more difficult to scale to 100GbE and higher.

FPGAs Advance Data Acceleration in the Digital Transformation Age

Applications in this new world such as AR/VR, autonomous vehicles, AI/ML and many others are no longer something in the future.  Achronix provides critical FPGA technology in the form of standalone FPGAs and embedded FPGA (eFPGA) IP, enabling this much needed data acceleration.