Gferasure is a high performance Galois field library for erasure coding and
algebraic signature computation. Gferasure implements constant by region
multiplication for erasure coding applications and dot product multiplication
for algebraic signatures along with normal Galois field multiplication.
Gferasure utilizes Intel's Single Instruction Multiple Data (SIMD) instruction sets Supplemental Streaming SIMD Extensions 3 (SSSE3), Streaming SIMD Extensions 4 (SSE4), and Advanced Vector eXtentions 2 (AVX2) for Intel processors along with NEON instructions for ARM processors. Gferasure currently supports template implementations for field sizes GF(2^8), GF(2^16), GF(2^32), and GF(2^64).
Building / Running
Information for building gferasure can be found in doc/build.md or doc/INSTALL
Technical Report: To be published in near future
We are happy to expand Gferasure, to contribute please fork and submit a pull request for review.
Nightly Build Tests
We host an archive of our nightly tests at is updated nightly at 2am Pacific.