Download the latest versions of Achronix application notes, datasheets, product briefs, user guides and white papers.
Select the individual tabs below to browse through each type of documentation. Or use the filter to only see documentation related to your product of interest.
Some documents are restricted (denoted by the lock symbol in the download button) and require a support portal account to access the download. To download a restricted document, enter your support portal account credentials when prompted. Don't have a support portal account? Register for an account here: Achronix Support Account Registration
FPGAs and eFPGAs Accelerate ML Inference at the Edge (WP026)
With the rapid proliferation of Internet-of-Things (IoT) and billions of connected devices, there is a paradigm shift taking place where big data is not only being processed in the core data center but also at the network edge. Field Programmable Gate Arrays (FPGAs), sitting at the intersection of performance and flexibility, are a promising solution for deep learning edge inference applications.
Data Orchestration Supports the Next Advance in AI (WP025)
Artificial intelligence (AI) and machine learning (ML) technologies now power a rapidly expanding range of product and applications from deeply embedded systems to hyperscale data-center deployments. Although there is a huge degree of diversity in the hardware designs supporting these applications, all require hardware acceleration. Data orchestration encompasses the pre- and post-processing operations that ensure the data seen by a machine learning engine arrives at an optimal speed and in the most suitable form for efficient processing.
An FPGA-Based Solution for a Graph Neural Network Accelerator (WP024)
Thanks to the rise of big data and the rapid increase in computing power, machine learning technology has experienced revolutionary development in recent years. Machine learning tasks such as image classification, speech recognition, and natural language processing, operate on Euclidean data with a certain size, dimension, and an orderly arrangement. However, in many realistic scenarios, data is represented by complex non-Euclidean data such as graphs. In this context, many new graph-based machine learning algorithm, or graph neural networks (GNNs), are constantly emerging in academia and industry.
The AI Evolution Calls for Adaptable Inferencing Platforms (WP023)
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.
FPGAs for Advanced Video Processing Solutions (WP022)
While the performance of an ASIC is typically high enough for broadcast-quality video processing, it supports only the feature set conceived of at design time and is not field upgradable. A CPU is the most flexible and easiest to design; however, clock frequencies have plateaued, and the era of dramatic improvements in performance are over. FPGAs represent a good balance between performance and flexibility for this class of applications.
FPGAs Enable the Next Generation of Communication and Networking Solutions (WP021)
Communications and networking systems that extend from the edges of the 5G network to the switches inside data centers are placing extreme pressure on the ability of silicon to support the computational and data-transfer rates they require. Traditionally programmable logic provided the best mixture of flexibility and speed for these systems, but has been challenged in recent years by the increase in speed. Through the inclusion of an innovative, multilevel network on chip that allows data to be streamed easily around the device without impacting the FPGA fabric, the Speedster7t architecture ensures all device resources are used to their full potential.
Eight Benefits of Using an FPGA with an On-chip High-Speed Network (WP020)
Since the initial introduction of FPGAs decades ago, each new architecture has continued to employ a bit-wise routing structure. While this approach has been successful, the rise of high-speed communication standards has required ever increasing on-chip bus widths to be able to support these new data rates. Achronix's solution was to create a revolutionary 2D high-speed network on chip (NoC) on top of the traditional segmented FPGA routing structure for its new Speedster7t FPGA family.
How to Design SmartNICs Using FPGAs to Increase Server Compute Capacity (WP017)
Intelligent server adapters, or SmartNICs, boost server performance in cloud and private data centers by offloading network processing workloads and tasks from server CPUs. Offloading network processing to a SmartNIC is not a new concept — for example, there are NICs that offload some network-processing functions such as checksum computation and segmentation. However, the rapid explosion in data-center network traffic driven by software-defined networking (SDN), Open vSwitch (OVS), and network functions virtualization (NFV) demands a new class of NIC with even greater offload capabilities: the SmartNIC.
Many users transitioning to Achronix eFPGA technology will be familiar with existing FPGA solutions from other vendors. Although Achronix technology and tools are similar to existing FPGA technology and tools, there are some differences. Understanding these differences are needed to achieve the very best performance and quality of results (QoR).
This tutorial serves as an introduce to the ACE engineering change order (ECO) suite — a set of Tcl commands that can add or remove instances, nets, pin connections, and more from a placed-and-routed design.
A Speedcore instance hosted in an SoC supports three different configuration modes: CPU, serial flash and JTAG. In CPU mode, an external CPU acts as the master and controls the programming operations for the Speedcore eFPGA, and offers a high-speed method for loading configuration data.
One of the desired requirements of any FPGA design tool is the ability to reproduce the exact same results every time the tool is run under the same conditions — a requirement refereed to as repeatability. The ACE placer and router are deterministic, delivering 100% repeatability.
When calculating dynamic power for a design, one input to any power estimation is the toggle rate of the signals. In most circumstances, the value used will be one of the industry standards of either 12.5% or 25% — values derived from a wide range of designs.
This application note covers the formal verification support available in the ACE environment. ACE currently is capable of supporting formal equivalency checking in its design flow, enabling the user to verify the synthesized netlist against the output at the different stages in the ACE flow.
In FPGA design, reset signals can sometimes have a significant effect on the overall quality of timing or routing results. Generally it is recommended to reduce the number of logic elements that need to be reset by taking advantage of initial values and coding in such a way that reset is only needed on a few end points.
Clock Design Planning for Speedcore eFPGAs (AN011)
Speedcore eFPGAs have a robust clocking architecture. While some designs only use a single main clock, others can have complicated clocking schemes. It is important for designers to understand the different types of clocks available in the Speedcore architecture, and how to get the best design out of the clocking resources available.
Maximize Hardware Assurance Using Embedded FPGAs (PB035)
Implementing a secure IP solution when developing a custom ASIC involves overcoming many risks along the development, manufacturing and supply chain flow. Hardware assurance continues to become more critical for military and defense applications as worldwide threats increase. By using an eFPGA IP solution to store mission critical IP, supply chain security is greatly simplified compared to the traditional ASIC design flow.
Speedcore eFPGA Test Chip Evaluation Board (PB030)
The Speedcore eFPGA evaluation board from Achronix contains the 16-nm Speedcore eFPGA test chip. The evaluation board’s Speedcore test chip has been customized with the right blend of resources such as LUTs, BRAMs, DSP64s, DFFs and a number of I/O so as to provide an optimum programmable platform for demonstrating, evaluating and testing Achronix’s Speedcore technology.
During normal SoC operation, the Speedcore eFPGA core requires configuration by the end user. This guide covers the details of how to configure a Speedcore instance via JTAG, CPU, or serial flash interface. Also included are details on the Achronix Configuration Bus (ACB) interface that can be used to program configuration bits for ASIC IP surrounding the Speedcore eFPGA.
Speedcore ASIC Integration and Timing User Guide (UG064)
This guide details the design flow for integrating a Speedcore eFPGA into an ASIC, including closing timing across the boundary between the Speedcore instance and the surrounding host ASIC, along with how to perform full-chip simulation.
Design for test (DFT) is an important consideration for Speedcore eFPGAs from the perspective of both Achronix and the ASIC integrator. The programmable nature of Speedcore eFPGAs deliver the inherent benefit of being able to use the programmable logic fabric to test itself. This guide describes Speedcore eFPGAs from a testability perspective and outlines the general features and methodologies that Achronix uses to achieve the necessary coverage.
This user guide lists each of the I/O pin groups available in the Speedster7t 7t1500 device, their functionality and recommended connection guidelines.
This product guide describes the function and operation of the Achronix Speedster7t FPGA SerDes for multi-standard applications and custom configurations.