Skip to content

The project uses a PCIe-Card for FPGA based zlib acceleration. This project provides a hardware accelerated version of zLib based compression/decompression RFC1950/RFC1951/RFC1952 with help of an FPGA based PCIe card.

License

Notifications You must be signed in to change notification settings

abalib/genwqe-user

 
 

Repository files navigation

genwqe-user

GenWQE (Generic Work Queue Engine) software supports the IBM PCIe3 FPGA or CAPI Compression Accelerator Adapters to speed up processing of the DEFLATE compression algorithm. This repository contains the source code to test, maintain and update the GenWQE PCIe/CAPI cards. Furthermore a zlib version with hardware acceleration is provided to do zlib style compression/decompression according to RFC1950, RFC1951 and RFC1952. This can be used as alternative to the traditional software zlib.

The GenWQE PCIe or CAPI cards can currently be acquired as options to the latest IBM System p machines:

And here the CAPI version of the adapter:

If you like to contribute to this project, please fill out and sign one of our contributor license agreements to be found in /licenses and send this back to us before sending us contributions.

Additional documentation can be found at the IBM Knowledgecenter.

A programming and usage guide for the hardware accelerated zlib can be downloaded here: Generic Work Queue Engine (GenWQE) Application Programming Guide.

The User's guide for the CAPI Compression Accelerator Adapter (FC EJ1A and EJ1B; CCIN 2CF0) can be found here: CAPI accelerated GZIP Compression Adapter User’s guide.

Possible distributors: If you take a snapshot of this into one of your releases, please let us know such that we can sync up testing. Thanks.

About

The project uses a PCIe-Card for FPGA based zlib acceleration. This project provides a hardware accelerated version of zLib based compression/decompression RFC1950/RFC1951/RFC1952 with help of an FPGA based PCIe card.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • C 89.5%
  • Shell 7.3%
  • Makefile 3.2%