Benchmarking nearest neighbors

Doing fast searching of nearest neighbors in high dimensional spaces is an increasingly important problem, but so far there has not been a lot of empirical attempts at comparing approaches in an objective way.

This project contains some tools to benchmark various implementations of approximate nearest neighbor (ANN) search for different metrics. We have pregenerated datasets (in HDF5) formats and we also have Docker containers for each algorithm. There's a test suite that makes sure every algorithm works.

Evaluated

• Annoy
• FLANN
• scikit-learn: LSHForest, KDTree, BallTree
• Weaviate
• PANNS
• NearPy
• KGraph
• NMSLIB (Non-Metric Space Library) : SWGraph, HNSW, BallTree, MPLSH
• hnswlib (a part of nmslib project)
• RPForest
• FAISS
• DolphinnPy
• Datasketch
• PyNNDescent
• MRPT
• NGT : ONNG, PANNG, QG
• SPTAG
• PUFFINN
• N2
• ScaNN
• Vearch
• Elasticsearch : HNSW
• Elastiknn
• OpenSearch KNN
• DiskANN : Vamana, Vamana-PQ
• Vespa
• scipy: cKDTree
• vald
• Qdrant
• qsgngtlib
• Milvus : Knowhere Glass
• pgvector

Data sets

We have a number of precomputed data sets for this. All data sets are pre-split into train/test and come with ground truth data in the form of the top 100 neighbors. We store them in a HDF5 format:

Results

These are all as of April 2023, running all benchmarks on a r6.16xlarge machine on AWS with --parallelism 31 and hyperthreading disabled. All benchmarks are single-CPU.

glove-100-angular

sift-128-euclidean

fashion-mnist-784-euclidean

nytimes-256-angular

gist-960-euclidean

glove-25-angular

TODO: update plots on http://ann-benchmarks.com.

Install

The only prerequisite is Python (tested with 3.10.6) and Docker.

1. Clone the repo.
2. Run pip install -r requirements.txt.
3. Run python install.py to build all the libraries inside Docker containers (this can take a while, like 10-30 minutes).

Running

1. Run python run.py (this can take an extremely long time, potentially days)
2. Run python plot.py or python create_website.py to plot results.
3. Run python data_export.py --out res.csv to export all results into a csv file for additional post-processing.

You can customize the algorithms and datasets if you want to:

• Check that algos.yaml contains the parameter settings that you want to test
• To run experiments on SIFT, invoke python run.py --dataset glove-100-angular. See python run.py --help for more information on possible settings. Note that experiments can take a long time.
• To process the results, either use python plot.py --dataset glove-100-angular or python create_website.py. An example call: python create_website.py --plottype recall/time --latex --scatter --outputdir website/.

Including your algorithm

1. Add your algorithm into ann_benchmarks/algorithms by providing a small Python wrapper.
2. Add a Dockerfile in install/ for it
3. Add it to algos.yaml
4. Add it to .github/workflows/benchmarks.yml

Principles

• Everyone is welcome to submit pull requests with tweaks and changes to how each library is being used.
• In particular: if you are the author of any of these libraries, and you think the benchmark can be improved, consider making the improvement and submitting a pull request.
• This is meant to be an ongoing project and represent the current state.
• Make everything easy to replicate, including installing and preparing the datasets.
• Try many different values of parameters for each library and ignore the points that are not on the precision-performance frontier.
• High-dimensional datasets with approximately 100-1000 dimensions. This is challenging but also realistic. Not more than 1000 dimensions because those problems should probably be solved by doing dimensionality reduction separately.
• Single queries are used by default. ANN-Benchmarks enforces that only one CPU is saturated during experimentation, i.e., no multi-threading. A batch mode is available that provides all queries to the implementations at once. Add the flag --batch to run.py and plot.py to enable batch mode.
• Avoid extremely costly index building (more than several hours).
• Focus on datasets that fit in RAM. For billion-scale benchmarks, see the related big-ann-benchmarks project.
• We mainly support CPU-based ANN algorithms. GPU support exists for FAISS, but it has to be compiled with GPU support locally and experiments must be run using the flags --local --batch.
• Do proper train/test set of index data and query points.
• Note that we consider that set similarity datasets are sparse and thus we pass a sorted array of integers to algorithms to represent the set of each user.

Authors

Built by Erik Bernhardsson with significant contributions from Martin Aumüller and Alexander Faithfull.

Related Publication

The following publication details design principles behind the benchmarking framework:

• M. Aumüller, E. Bernhardsson, A. Faithfull: ANN-Benchmarks: A Benchmarking Tool for Approximate Nearest Neighbor Algorithms. Information Systems 2019. DOI: 10.1016/j.is.2019.02.006

Related Projects

• big-ann-benchmarks is a benchmarking effort for billion-scale approximate nearest neighbor search as part of the NeurIPS'21 Competition track.