Turbocharging Vector Databases using Modern SSDs
Summary: SSD-aware HNSW optimizations for pgvector combine io_uring parallel I/O, spatially ordered insertion, and locality-preserving colocation. They deliver up to 11.1× higher throughput, 3.23× better cache hit ratio, and 98.4% faster index builds. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Joobo Shim (Seoul National University)
- 2. Jaewon Oh (Seoul National University)
- 3. Hongchan Roh (Dnotitia)
- 4. Jaeyoung Do (Seoul National University)
- 5. Sang-Won Lee (Seoul National University)
BibTeX Citation
@article{shim_vldb25,
title = {{Turbocharging Vector Databases using Modern SSDs}},
author = {Shim, Joobo and Oh, Jaewon and Roh, Hongchan and Do, Jaeyoung and Lee, Sang-Won},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {11},
pages = {4710--4722},
doi = {10.14778/3749646.3749724},
url = {https://doi.org/10.14778/3749646.3749724},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 1 of 1 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,209 | CMANNS: GPU-Accelerated Graph Index Construction for ANNS via Compute-Memory Disaggregation | 2026 | SIGMOD | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 914 | A Modeling Study of the TPC-C Benchmark | 1993 | SIGMOD | 0.00013246456 |
| 1,760 | Starling: An I/O-Efficient Disk-Resident Graph Index Framework for High-Dimensional Vector Similarity Search on Data Segment | 2024 | SIGMOD | 9.8157819e-05 |
Previous
Page 1 / 1
Next