A production-proven Oracle sorting algorithm with claimed 5× performance gains is now public domain, enabling MySQL, PostgreSQL, and other open source databases to adopt it freely and accelerate performance without architectural changes.
An expired Oracle patent has opened the door for open source databases to freely adopt a high-performance sorting algorithm that was previously locked behind intellectual property restrictions. Patent US7680791B2, which covers a fast method for sorting database keys, has now entered the public domain, removing licensing barriers for open-source implementations.
The algorithm, informally known as “Orasort”, was invented by Mark Callaghan during his time at Oracle and originally shipped with Oracle Database 10gR2. It was designed to accelerate sorting workloads involving similar keys with shared prefixes—a common scenario in database systems.
Orasort introduces several optimisations, including skipping repeated comparisons of common prefix bytes, adaptively switching between quicksort and radix sort, caching key substrings to reduce CPU cache misses, prefetching data to minimise wasted memory access, and producing partial results before the entire sort completes to reduce latency.
According to Callaghan, the algorithm delivered approximately five times faster in-memory sorting compared to Oracle’s earlier approach and compared favourably with commercial alternatives. “The new sort was often about five times faster than the old sort,” he said, noting its strong performance inside Oracle’s DBMS.
With the patent expired, the database community has shown immediate interest in implementing Orasort in MySQL, PostgreSQL, and other open-source engines. The unusually clear and readable patent documentation further lowers adoption friction. Early experimentation has already begun, with Hannu Krosing, a database engineer at Google, attempting implementations in Python, C, and C++ using AI tools such as Gemini.
While Oracle continues to hold more than 52,000 patents, this expiration stands out for releasing a core database performance primitive rather than a peripheral feature, creating new opportunities for open-source innovation.
