SplinterDB: Closing the Bandwidth Gap on NVMe
, Rutgers University
Date: Monday, October 21, 2019
Time: 2:00 PM to 3:00 PM Note: all times are in the Eastern Time Zone
Location: G32-449 (Patil/Kiva)
Event Type: Seminar
Room Description: G32-449 (Patil/Kiva)
Host: Julian Shun, MIT CSAIL
Contact: Julian J. Shun, email@example.com
Relevant URL: http://fast-code.csail.mit.edu/
Speaker URL: https://ajhconway.com/
firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
TALK: SplinterDB: Closing the Bandwidth Gap on NVMe
Abstract: Key-value stores are a fundamental component to storage systems. Recent advances in storage hardware, especially the advent of NVMe devices, expose new bottlenecks in the design of current key-value stores. We introduce SplinterDB and show how it achieves a 6-9x improvement in the insertion rate over Facebook's RocksDB on NVMe devices, while also improving query performance by 2-2.5x.
Bio: Alex Conway is a PhD student in Computer Science at Rutgers University, advised by Martin Farach-Colton. His research is focused on external memory theory and storage systems. His recent publications include work on high performance key-value stores, file system aging, external memory hash tables and the analysis of storage optimization heuristics.
He is a member of the BetrFS team. BetrFS, a prototype file system, is designed around the principles of write-optimization and is built using variants of B\epsilon-trees.
Algorithms & Theory, Systems & Networking
Created by Julian J. Shun at Sunday, October 13, 2019 at 4:57 PM.