Skip to content

Add skim: ultra-fast, high-throughput compression algorithm #255

Description

@VantorreWannes

I would like to propose adding skim to the Squash benchmark suite.

skim is an open-source, ultra-fast compression algorithm designed specifically for high-throughput scenarios where execution speed is prioritized over compression ratio. It occupies a unique point on the Pareto frontier, sitting between density and raw memcpy.

Silesia Benchmark Reference (Ryzen 7 7800X3D)

  • Compression Speed: ~3.39 GB/s
  • Decompression Speed: ~5.96 GB/s
  • Ratio: ~64.5%

Integration Details

  • Repository: https://github.com/VantorreWannes/skim
  • License: MIT
  • API: Written in Zig, but exposes a clean C API (simple state initialization, compression, and decompression functions), making it straightforward to wrap for Squash's C-based plugin architecture.

Adding skim would provide Squash users with a valuable addition in the extremely low-latency/high-speed compression space.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions