ParlayANN: Scalable and Deterministic Parallel Graph-Based Approximate Nearest Neighbor Search Algorithms
Speaker
Host
Julian Shun
CSAIL MIT
Abstract: Approximate nearest-neighbor search (ANNS) algorithms are a key part of the modern deep learning stack due to enabling efficient similarity search over high-dimensional vector space representations (i.e., embeddings) of data. Among various ANNS algorithms, graph-based algorithms are known to achieve the best throughput-recall tradeoffs. Despite the large scale of modern ANNS datasets, existing parallel graph based implementations suffer from significant challenges to scale to large datasets due to heavy use of locks and other sequential bottlenecks, which 1) prevents them from efficiently scaling to a large number of processors, and 2) results in nondeterminism that is undesirable in certain applications.
In this paper, we introduce ParlayANN, a library of deterministic and parallel graph-based approximate nearest neighbor search algorithms, along with a set of useful tools for developing such algorithms. In this library, we develop novel parallel implementations for four state-of-the-art graph-based ANNS algorithms that scale to billion-scale datasets. Our algorithms are deterministic and achieve high scalability across a diverse set of challenging datasets. In addition to the new algorithmic ideas, we also conduct a detailed experimental study of our new algorithms as well as two existing non-graph approaches. Our experimental results both validate the effectiveness of our new techniques, and lead to a comprehensive comparison among ANNS algorithms on large scale datasets with a list of interesting findings. This work is joint with Zheqi Shen, Guy Blelloch, Laxman Dhulipala, Yan Gu, Harsha Vardhan Simhadri, and Yihan Sun and appeared in PPoPP 2024.
Bio: Magdalen Dobson Manohar is a 5th year PhD student in the Computer Science Department at Carnegie Mellon University advised by Guy Blelloch. She is interested in designing parallel and concurrent algorithms for solving problems related to similarity search, information retrieval, and computing nearest neighbors, with a particular focus on similarity search in high dimensions. She will be joining Microsoft as a Senior Researcher in Summer 2024. She completed her undergraduate degree in mathematics at MIT in Spring 2019.
In this paper, we introduce ParlayANN, a library of deterministic and parallel graph-based approximate nearest neighbor search algorithms, along with a set of useful tools for developing such algorithms. In this library, we develop novel parallel implementations for four state-of-the-art graph-based ANNS algorithms that scale to billion-scale datasets. Our algorithms are deterministic and achieve high scalability across a diverse set of challenging datasets. In addition to the new algorithmic ideas, we also conduct a detailed experimental study of our new algorithms as well as two existing non-graph approaches. Our experimental results both validate the effectiveness of our new techniques, and lead to a comprehensive comparison among ANNS algorithms on large scale datasets with a list of interesting findings. This work is joint with Zheqi Shen, Guy Blelloch, Laxman Dhulipala, Yan Gu, Harsha Vardhan Simhadri, and Yihan Sun and appeared in PPoPP 2024.
Bio: Magdalen Dobson Manohar is a 5th year PhD student in the Computer Science Department at Carnegie Mellon University advised by Guy Blelloch. She is interested in designing parallel and concurrent algorithms for solving problems related to similarity search, information retrieval, and computing nearest neighbors, with a particular focus on similarity search in high dimensions. She will be joining Microsoft as a Senior Researcher in Summer 2024. She completed her undergraduate degree in mathematics at MIT in Spring 2019.