Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the next generation of data compression technology in this conference talk that introduces OpenZL, a revolutionary approach to lossless compression that moves beyond traditional algorithmic improvements. Discover how OpenZL models a compressor as a dynamically configurable graph of proven building blocks, achieving compression ratios that match or surpass domain-specific codecs while maintaining higher throughput through a single universal decoder and unified, security-hardened codebase. Learn about the innovative design and graph architecture, including how the optimizer assembles blocks into efficient, dataset-specific compressors tailored for various data types. Examine comprehensive benchmark results comparing OpenZL's performance against Zstandard and specialized codecs across columnar data, serialization schemas, floating point numbers, and binary-blob workloads, demonstrating significant improvements in both compression ratio and processing speed. Understand the operational advantages of this unified approach, including simplified deployment with one package to ship, flexible configuration options, support for private extensions, and streamlined auditing and maintenance processes. Gain insights into the open specification roadmap and learn how Linux Foundation members can contribute to shaping the 1.0 release of this groundbreaking compression technology that promises to redefine the landscape of general-purpose lossless compression after a decade of incremental improvements.
