Rowles.LeanLucene.Search.Simd
Classes
SimdIntersection
Specialised SIMD-accelerated Boolean AND intersection for sorted doc-ID arrays.
SimdIntrinsicsVectorOps
Explicit-intrinsics SIMD path for vector arithmetic. Selects the widest available instruction set at runtime: AVX-512 → AVX2 → SSE → scalar. Provided alongside SimdVectorOps for benchmark comparison; the slower of the two will be removed once measurements are concluded.
SimdVectorOps
SIMD vector primitives used by HNSW graph operations and exact rerank. Built around System.Numerics.Vector`1, which the .NET JIT lowers to AVX2 or AVX-512 instructions automatically on capable hardware.