DATE:Thursday, October 13, 2022
TIME:12:00 - 1:00 pm
PLACE: RMCIC Building, 4th Floor Panther Hollow Conference Room (4105)
A zoom link will be sent to the SDI mailing list a day before the in-person talk.
SPEAKER: Harsha Simhadri, Microsoft Research
TITLE: Vector Search Systems for Web-scale Search and Recommendation
ABSTRACT:
Web-scale search and recommendation scenarios increasingly use Vector search aka Approximate Nearest Neighbor Search (ANNS) indices to retrieve objects based on the similarity of their learnt representations in a geometric space. Since these scenarios often span billions or trillions of objects, efficient and scalable ANNS algorithms are critical to making these systems practical. However, most algorithms studied in literature either focus on small million-scale datasets or do not have features necessary for practical indices, e.g., external memory indices or support for real-time updates.
In this talk we discuss progress on this problem. Specifically, we present DiskANN, the first external memory ANNS algorithm that can index a billion points on a single commodity machine with <64GB DRAM and serve queries at few milliseconds latency. This represents an order of magnitude more points indexed per machine than previous work. In addition, the index allows real-time updates and its in-memory performance compares well with other state of the art indices. In a recently organized billion-scale ANNS challenge at NeurIPS'21, DiskANN proved to be the state of the art for many billion-scale datasets on both standard and specialized hardware. We will overview the datasets and baselines released in this challenge.
Finally, with a view towards making vector search the primary choice of index for search and recommendation scenarios, we will highlight some open problems. A few of these include support for hybrid queries that combine vector search and hard matches such as language or author; crash recovery and serializability; systems for extremely large distributed indices; support for out-of-distribution queries. We will share a few preliminary results in these directions.
Joint work with Ravishankar Krishnaswamy, Suhas J Subramanya, Aditi Singh, Rohan Kadekodi, Devvrit, Shikhar Jaiswal, Magdalen Dobson, Siddharth Gollapudi, Neel Karia, Varun Sivasankaran.
BIO:
Harsha Simhadri is a Principal Researcher at Microsoft Research. He enjoys developing new algorithms that enable practical systems. Recent examples include algorithms for web-scale nearest-neighbor search deployed in various Microsoft scenarios, and new ML operators and architectures for tiny IoT and edge devices. He previously worked on parallel algorithms and run-times with provable guarantees for multi-core processors for his PhD thesis at Carnegie Mellon University.
VISITOR HOST: Suhas Jayaram Subramanya
VISITOR COORDINATOR: Karen Lindenfelser
SDI SEMINAR QUESTIONS?
Karen Lindenfelser, 86716, or visit www.pdl.cmu.edu/SDI/