EvenDB: Optimizing key-value storage for spatial locality

Eran Gilad, Edward Bortnikov, Anastasia Braginsky, Yonatan Gottesman, Eshcar Hillel, Idit Keidar, Nurit Moscovici, Rana Shahout

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Applications of key-value (KV-)storage often exhibit high spatial locality, such as when many data items have identical composite key prefixes. This prevalent access pattern is underused by the ubiquitous LSM design underlying high-Throughput KV-stores today. We present EvenDB, a general-purpose persistent KV-store optimized for spatially-local workloads. EvenDB combines spatial data partitioning with LSM-like batch I/O. It achieves high throughput, ensures consistency under multi-Threaded access, and reduces write amplification. In experiments with real-world data from a large analytics platform, EvenDB outperforms the state-of-The-Art. E.g., on a 256GB production dataset, EvenDB ingests data 4.4X faster than RocksDB and reduces write amplification by nearly 4X. In traditional YCSB workloads lacking spatial locality, EvenDB is on par with RocksDB and significantly better than other open-source solutions we explored.

Original languageEnglish
Title of host publicationProceedings of the 15th European Conference on Computer Systems, EuroSys 2020
ISBN (Electronic)9781450368827
DOIs
StatePublished - 15 Apr 2020
Event15th European Conference on Computer Systems, EuroSys 2020 - Heraklion, Greece
Duration: 27 Apr 202030 Apr 2020

Publication series

NameProceedings of the 15th European Conference on Computer Systems, EuroSys 2020

Conference

Conference15th European Conference on Computer Systems, EuroSys 2020
Country/TerritoryGreece
CityHeraklion
Period27/04/2030/04/20

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Networks and Communications
  • Hardware and Architecture
  • Software
  • Control and Systems Engineering

Fingerprint

Dive into the research topics of 'EvenDB: Optimizing key-value storage for spatial locality'. Together they form a unique fingerprint.

Cite this