Database / Milvus Vector database Interview questions
What are the main vector index types Milvus supports?
Milvus supports several index families, each suited to different priorities around speed, memory usage, accuracy, and dataset scale.
| Index type | Best suited for |
| FLAT | Exact, brute-force search; small datasets or when 100% recall is required. |
| IVF_FLAT / IVF_SQ8 / IVF_PQ | Balancing speed, memory, and accuracy at larger scale via clustering and quantization. |
| HNSW | High recall and low query latency; a very common general-purpose default. |
| DiskANN | Very large datasets that don't fit comfortably in memory, using disk-based indexing. |
| SCANN | High-throughput search optimized for particular hardware/quantization trade-offs. |
| GPU indexes (e.g. GPU_CAGRA) | Hardware-accelerated search and index building on NVIDIA GPUs. |
There's no single universally-best index; the right choice depends on dataset size, available memory, whether GPUs are available, and the acceptable trade-off between recall (search accuracy) and latency for the specific application.
Which index type performs exact, brute-force search?
DiskANN is particularly well suited for:
More Related questions...
