PilotANN
PilotANN is a memory-bounded GPU-accelerated framework for large-scale vector search, designed to improve performance and efficiency of approximate nearest neighbor (ANN) search workloads, making it relevant as a high-performance engine/component in vector database and vector search systems.
About this tool
title: PilotANN slug: pilotann url: https://github.com/ytgui/PilotANN category: sdks-libraries featured: false brand: "" brand_logo_url: "/" tags:
- gpu-acceleration
- ann
- high-performance
summary: | PilotANN is a memory-bounded, GPU-accelerated framework for graph-based approximate nearest neighbor (ANN) vector search, intended as a high-performance engine or component for large-scale vector databases and vector search systems.
features:
- Memory-bounded GPU acceleration designed specifically for large-scale vector search workloads.
- Focus on graph-based approximate nearest neighbor (ANN) search.
- Suitable as a high-performance engine/component within vector databases and vector search pipelines.
- Implementation built on top of LibTorch (PyTorch C++ backend).
- Integrates with a FAISS (Facebook AI Similarity Search) build within the same Python environment.
- Python bindings via setup.py for development (
python3 ./setup.py develop). - Benchmark scripts for evaluating performance on multiple datasets (e.g.,
laion-1m,laion-100m). - Configurable benchmark parameters such as dataset choice, sample ratio, and principal dimension (
d_principle). - Support for multi-CPU execution with OpenMP thread limiting (
OMP_THREAD_LIMIT). - Example dataset directory structure for deep and LAION datasets (deep-1m, deep-100m, laion-100m).
technical_requirements:
- Linux environment (instructions provided for RHEL/CentOS-like systems).
- CUDA-capable NVIDIA GPU with CUDA Toolkit 11.8 and appropriate NVIDIA driver.
- Python environment (e.g., Anaconda/Miniconda) with Python 3.10.
- PyTorch 2.1.2 with CUDA 11.8, torchvision, torchaudio, and numpy.
- Build tools and libraries: CMake, Git, SWIG, LAPACK and LAPACK-devel, compiler toolchain (Development Tools group on RHEL/CentOS).
- FAISS 1.8.0 built with Python bindings (GPU disabled in the provided example build config).
usage:
- Build FAISS under the active Python environment, then build and develop-install PilotANN via
python3 ./setup.py develop. - Prepare datasets under a
.datasets/directory following the documented structure. - Run benchmarks using
script/bench_1.pywith options for dataset name, sample ratio, and principal dimension; control CPU parallelism viaOMP_THREAD_LIMIT.
pricing: | PilotANN is an open-source project hosted on GitHub; no pricing information or paid plans are specified in the repository content.
notes: | The repository README focuses on build, environment, and benchmarking instructions. Detailed internal algorithm descriptions, API reference, and configuration options beyond the benchmark script parameters are not included in the provided content.
Loading more......
Information
Categories
Tags
Similar Products
6 result(s)NVIDIA CAGRA is a GPU-accelerated graph-based library for approximate nearest neighbor searches, optimized for high-performance vector search leveraging modern GPU parallelism. It is suitable for scenarios requiring rapid, large-scale vector retrieval.
BANG is a billion-scale approximate nearest neighbor search system optimized for single GPU execution, enabling high-performance vector search in vector database environments at massive scale.
EFANNA is an extremely fast approximate nearest neighbor search algorithm based on kNN graphs and randomized KD-trees. The provided implementation offers a high-performance ANN index suitable as a building block in custom vector search and retrieval infrastructure.
jvector is a high-performance Java-based library and engine for vector search and approximate nearest neighbor indexing.
RTNN is a research prototype system and codebase that accelerates high-dimensional nearest neighbor search using hardware ray tracing units on modern GPUs. It targets vector similarity search workloads common in AI applications, exploring ray-tracing hardware as an alternative acceleration path to traditional CPU- or CUDA-based ANN indexes.
DiskANN is a graph-based approximate nearest neighbor search (ANNS) system optimized for fast and accurate billion-point nearest neighbor search on a single node, leveraging SSD storage. It is highly relevant for large-scale vector database applications requiring efficient vector search at scale.