Intel Releases x86-simd-sort 5.0 With 4~5x Faster C++ Object Sorting Using AVX-512

It's been nearly one year to the day since outlining intel's AVX-512 powered sorting library to offer blazing fast sort speeds. Over the past year has brought the 1.0 release, new algorithms in v2.0, AVX2 support and more AVX-512 optimizations in v4.0, and now today Intel is out with x86-simd-sort 5.0 with yet more performance improvements.

The x86-simd-sort 5.0 release introduces a new API to sort custom-defined C++ objects via object_qsort. This new sort support of custom C++ objects can be 4~5x faster than using std::sort on AVX-512 systems but ultimately the performance impact will vary based upon the custom class.

The x86-simd-sort 5.0 release also adds a new keyvalue_qsort API to sort arrays representing key-value pairs and is much faster with this new API. The v5.0 release also adds AVX2 support for the argosrt and argselect methods. Those AVX2 additions have already landed into upstream NumPy for NumPy 2.0 with that Python library having been one of the initial adopters of this high performance Intel library.

Downloads and more details on Intel's x86-simd-sort 5.0 release via GitHub.