Aug 26, 2026 · Artificial Intelligence

How Alibaba Cloud Elasticsearch’s Cloud‑Native Vector Engine Tops VectorDBBench

Alibaba Cloud Elasticsearch on ES 9.4, using the FalconSeek HNSW engine, achieves 82,520 QPS at 0.98 recall with a 1.8 ms P99 latency in VectorDBBench, and the article explains the end‑to‑end architectural redesign—including quantized candidate recall, batch distance computation, hot‑data layout, on‑demand re‑ranking, and segment lifecycle integration—that makes these results possible.

AI SearchCloud NativeElasticSearch

0 likes · 17 min read

How Alibaba Cloud Elasticsearch’s Cloud‑Native Vector Engine Tops VectorDBBench

ByteDance SE Lab

Jul 8, 2026 · Databases

Volcano Milvus Hits Nearly 3× VectorDBBench Leader in Retrieval Speed

Under a fixed monthly budget of about 7,100 CNY, Volcano Milvus combines DiskANN and RaBitQ (including Extended‑RaBitQ) with in‑memory layout and query‑path slimming to deliver 20,420 QPS, 2.5 ms average latency, 93.9 % recall, achieving nearly three times the VectorDBBench top score while using roughly one‑third of the memory, disk and compute resources of competing solutions.

DiskANNMilvusPerformance Benchmark

0 likes · 12 min read

Volcano Milvus Hits Nearly 3× VectorDBBench Leader in Retrieval Speed