Example of Integrating KBest into Milvus

The KBest algorithm can be integrated into the Milvus database (version 2.4.5) to accelerate the query efficiency while ensuring a high recall rate.

Among all index algorithms supported by Milvus, the graph-based index algorithm is Hierarchical Navigable Small World (HNSW), which can perform quick query and achieve a high recall rate, but consumes a large amount of memory resources. To extend the graph-based index algorithm and accelerate the query while ensuring a high recall rate, the KBest algorithm optimizes the performance and precision of the nearest neighbor search by using methods such as quantization and vector instruction, providing the search capability equivalent to the open source Faiss HNSW algorithm. The procedure is as follows:

1. Install Milvus.

   For details, see the Milvus Database Installation Guide.

2. Apply the patch file into Milvus for full compilation.

   For details, see the Milvus Database KBest Optimization Feature Guide.

3. Use ann-benchmarks for the test.

   Perform the test by following instructions in the Milvus Database ANN-Benchmarks Test Guide.