# intel/isa-l

Intelligent Storage Acceleration Library

Repository: https://github.com/intel/isa-l
Canonical: https://ross.abutalabs.com/products/isa-l
Language: C
License: NOASSERTION
License Family: other
Last push: 2026-07-22T09:42:25+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 93, release rhythm 59, longevity 100
- inputs: {"age_days": 3873, "days_push": 42, "days_rel": 63, "gap_med": 268.5, "n_releases_24m": 3}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1093, forks 344 (observed 2026-08-28T04:03:33.734816+00:00)

## What it is
Intel's Intelligent Storage Acceleration Library (ISA-L) is a C library of SIMD-optimized low-level functions for storage applications, including erasure codes, CRC variants, RAID parity, and fast deflate/inflate compression. It also ships igzip, a gzip-like command-line compression tool accelerated by the library.

## Use cases
- fast erasure coding for distributed storage
- compute CRC32/CRC64 checksums quickly
- calculate RAID XOR and P+Q parity
- deflate-compatible compression with high throughput
- accelerate gzip-style compression on x86 or aarch64
- add Reed-Solomon encoding/decoding to a storage system

## When to choose
- you need maximum-throughput erasure coding, CRC, RAID parity, or deflate compression in C/C++
- you target x86_64, aarch64, or RISC-V and want SIMD-accelerated primitives
- you want a battle-tested Intel-maintained storage acceleration library

## When to avoid
- you need fully validated, safe APIs - parameters are not checked and invalid input causes undefined behavior
- you need a high-level or pure-portable solution without assembly dependencies like nasm
- you need cryptographic hashing - use isa-l_crypto instead

## Facets
- artifact type: library
- maturity: stable
- function: compression, cryptography, math
- domain: developer-tools, performance, files
- platform: windows, cpp, cross-platform
- tags: erasure-coding, crc, raid, simd, deflate, igzip, storage-acceleration, linux, macos

## Member repositories
- intel/isa-l (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:33.734816+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:47:50.665568+00:00, confidence not recorded.
  - readme: https://github.com/intel/isa-l (fetched 2026-08-28T04:03:33.734816+00:00, sha c43bf30b8f4e)
- Data as of 2026-08-30T08:39:29.467469+00:00.