AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

SPANN: Highly-efficient Billion-scale Approximate Nearest Neighborhood Search

Qi Chen, Bing Zhao, Haidong Wang, Mingqin Li, Chuanjie Liu, Zengzhong Li, Mao Yang, Jingdong Wang

NeurIPS 2021 · Architectures and efficiency › RNNs and normalization · approach: Distributed LLM inference systems

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Hierarchical balanced clustering for centroids instead of random or hierarchical KMeans (ablation)SPANN · SIFT1M recall@1 and recall@10recall vs latencyImproved
Increase the number of centroids toward 16% of points (ablation)SPANN · SIFT1Mrecall vs latencyImproved
Increase centroids beyond 16% of points (ablation)SPANN · SIFT1Mrecall vs latencyNo clear effect
Use more than one closure replica per vector (ablation)SPANN · SIFT1Mrecall vs latencyImproved
Increase closure replicas above 8 (ablation)SPANN · SIFT1Mrecall vs latencyNo clear effect
Add query-aware dynamic pruning of posting lists (ablation)SPANN · SIFT1Mquery latency at same recallImproved
Replace random partition with multi-constraint balanced clustering (ablation)Distributed SPANN, 32 partitions · SPACEV1Bmachines dispatched per query: 32 -> 9Improved
Add closure assignment to balanced clustering (ablation)Distributed SPANN, 32 partitions · SPACEV1Bmachines dispatched per query: 9 -> 8Improved
Add query-aware dynamic pruning online (ablation)Distributed SPANN, 32 partitions · SPACEV1Bmachines dispatched per query: 8 -> 6.3Improved
Closure assignment increases vectors stored per machineDistributed SPANN · SPACEV1B, 8/16/32 partitionsvectors per machine: 20%Hurt
Use more machines (8, 16, 32 partitions) with balanced partitioning vs random partitionDistributed SPANN · SPACEV1B query accesses per machinequery accesses per machineImproved
SPANN instead of in-memory graph ANNS (NSG, HNSW, SCANN, NGT, N2)SIFT1M · SIFT1M recall@1, recall@10VQ capacityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy