Fletcher: A framework to integrate FPGA accelerators with Apache Arrow
Fletcher is a framework that helps to integrate FPGA accelerators with tools that use Apache Arrow in their back-ends.
While many software projects can benefit from these advantages, also hardware accelerated applications have seen serious serialization bottlenecks. Fletcher focuses on FPGA accelerators only. Through Fletcher and Arrow, efficient FPGA acceleration is made available to all the supported languages.
(If you are looking for GPGPU's, some work is being done as part of the Arrow project already, go check out the Arrow repository.)
Given an Arrow Schema (description of a tabular datastructure), Fletcher generates the following:
- An easy-to-use hardware interface for the functional part of the accelerator
- You provide a range of table indices (rather than byte address)
- You receive streams of the datatype specified in the schema (rather than bus words)
- No pointer arithmetic, reordering, parallelizing or serializing, buffering, etc.. required!
- A template for the functional part of accelerator (to be implemented manually or with HLS)
Our framework is functional, but in early stages of development. It currently supports the following features:
Apache Arrow support:
- Reading/writing an Arrow Schema created from any combination of:
- Fixed-width primitives
- Validity bitmaps are supported.
- Not supported (yet):
- Our core hardware descriptions are vendor independent; we don't use any vendor IP.
- There is an example project for the Amazon EC2 F1 and POWER8/9 CAPI SNAP FPGA platforms.
- Our bus interconnect is similar to AXI, so it should be easy to integrate in many existing platforms.
- Upcoming: OpenCAPI support
- How to generate a Column Reader.
- For the current state of the project, this would be your starting point in hardware design.
- How to use the host-side run-time library.
- Regular Expression matching example on Amazon EC2 F1
- Regular Expression matching example on CAPI SNAP