English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Approximate nearest neighbor search (ANNS) plays an indispensable role in awide variety of applications, including recommendation systems, informationretrieval, and semantic search. Among the cutting-edge ANNS algorithms,graph-based approaches provide superior accuracy and scalability on massivedatasets. However, the best-performing graph-based ANN search solutions incurtens of hundreds of memory footprints as well as costly distance computation,thus hindering their efficient deployment at scale. The 3D NAND flash isemerging as a promising device for data-intensive applications due to its highdensity and nonvolatility. In this work, we present the near-storage processing(NSP)-based ANNS solution Proxima, to accelerate graph-based ANNS withalgorithm-hardware co-design in 3D NAND flash. Proxima significantly reducesthe complexity of graph search by leveraging the distance approximation andearly termination. On top of the algorithmic enhancement, we implement Proximasearch algorithm in 3D NAND flash using the heterogeneous integrationtechnique. To maximize 3D NAND's bandwidth utilization, we present customizeddataflow and optimized data allocation scheme. Our evaluation results showthat: compared to graph ANNS on CPU and GPU, Proxima achieves a magnitudeimprovement in throughput or energy efficiency. Proxima yields 7x to 13xspeedup over existing ASIC designs. Furthermore, Proxima achieves a goodbalance between accuracy, efficiency and storage density compared to previousNSP-based accelerators.

Proxima: Near-storage Acceleration for Graph-based Approximate Nearest Neighbor Search in 3D NAND