Smaller and faster data compression with Zstandard

Introduced Zstandard (Zstd) 1.0, a new lossless compression algorithm and implementation designed for modern hardware. Zstd offers significant improvements over Deflate (gzip, zlib) in both compression and decompression speeds, while maintaining comparable or better compression ratios. Key technical innovations include a performance-first design optimized for modern CPUs, a scalable architecture with 22 compression levels, parallel execution capabilities through a new Huffman decoder (Huff0), and a branchless design to avoid costly pipeline flushes. Zstd's memory window is significantly larger than zlib's, allowing for better compression. The post details benchmark comparisons against lz4, xz, and zlib using the Silesia corpus and command-line tools, highlighting Zstd's superior trade-offs.
More from the Scalability and Large-Scale Systems Engineering trail
- Open-Sourcing Rebalancer: A Generic, High-Performance Library for Solving Assignment Problems
- ZGateway: Learnings from Putting a Proxy in Front of ZippyDB
- MTIA 300: Meta’s First Training Chip with Built-in NICs and Communication-Offloading Engines
- From User Sequences to Scaling Laws: A Multi-Stage Architecture for Meta’s Ads Ranking
- GEM Training: How Meta Doubled the Efficiency of Its LLM-Scale Ads Foundation Model