Intel recently announced its new Intel Open FPGA Stack (Intel OFS), a scalable, source-accessible hardware and software infrastructure delivered through git repositories. This makes it easier for hardware, software, and application developers to create custom acceleration platforms and solutions.
Additionally, Intel OFS provides standard interfaces and APIs to enable greater code reuse, accelerate development and speed deployment.
The challenge for any new FPGA-based acceleration platform development – comprised of FPGA hardware design, Intel Xeon Scalable processor-ready software stack and application workloads – centers on how much to develop from scratch versus reuse or license.
A high-level block diagram outlines the hardware and software deliverables for Intel Open FPGA Stack (Intel OFS). The bottom half depicts a PCI Express board with an Intel FPGA. Board and application developers use the hardware code to develop the FPGA interface manager and workloads. The Intel FPGA uses the PCI Express interface to pass data to and from the host beginning with the Linux drivers and open programmable acceleration engine library. Intel OFS also includes foundational application examples.
Intel OFS provides a customizable software and hardware infrastructure that satisfies many of the pain points for hardware, software, and application developers, including modular and “composable” code from which to develop the FPGA design (“take and tailor”), open-sourced and upstreamed code to the Linux kernel enabling open-source distribution vendors to provide native support to third parties and proprietary Intel-OFS platforms.
In short, the respective values to hardware, application and software developers are customization, easy portability across Intel FPGA platforms and native support in major OS vendor distributions.
“FPGAs have always and continue to enable developers to deliver customized hardware with optimal power, performance and TCO for workloads from the edge to the cloud” said Dave Moore, Intel corporate VP and general manager of the Programmable Solutions Group. “With the proven success from our early-access customers, we are excited to launch the Intel Open FPGA Stack, with its demonstrated ability to dramatically both reduce the development time and also increase code and hardware design reuse for customers and partners looking to accelerate their workloads.”
Now, board developers, original design manufacturers, and customers can leverage a common infrastructure with standard interfaces to jump-start their FPGA-hardware development. Application developers can achieve a better return on their developments with greater portability across different Intel OFS-based platforms.
Leading open-source software vendors can also respond to customers’ needs to offer expanded support for FPGAs as they do for CPUs and GPUs under existing and new contracts can now use Intel’s open-sourced and upstreamed code.
Developers interested in trying the new Intel OFS for their next project or would like details on the Early Access Program (EAP) should contact an Intel sales representative to get started. The EAP for Intel OFS runs through most of 2021.
Learn more here.
