Transcoding unicode characters with AVX‐512 instructions

Abstract Intel includes in its recent processors a powerful set of instructions capable processing 512‐bit registers with single instruction (AVX‐512). Some these have no equivalent earlier sets. We leverage to efficiently transcode strings between the most common formats: UTF‐8 and UTF‐16. With our novel algorithms, we are often twice as fast previous best solutions. For example, Chinese text from UTF‐16 at more than 5 GiB using fewer 2 CPU per character. To ensure reproducibility, make software freely available an open‐source library. Our library is part popular Node.js JavaScript runtime.