What question did this study set out to answer?

The aim is to develop a time-series database that is lightweight and easy to use for real-time data handling.

April 23, 2026Open Access

ZeitDB - A Single Node Time Series Database

VKVansh KumarNational Society of Professional Engineers VAVedant AryaNational Society of Professional Engineers SJSankalp JhanjeeNational Society of Professional Engineers

Key Points

The aim is to develop a time-series database that is lightweight and easy to use for real-time data handling.
Developed a single node database using an LSM-tree design and write-ahead log.
Implemented in-memory buffering and SSTable flushes for efficient data management.
Created tagging and alerting components for data filtering and anomaly detection.
ZeitDB effectively handles fast, timestamped data with reduced complexity.
Retention rules maintain manageable storage sizes without sacrificing performance.
The alerting component successfully identifies unusual data patterns and threshold breaches.

Abstract

ZeitDB is a lightweight time-series database built to make handling fast, timestamped data simple and practical. It uses an easy to understand LSM-tree design with a write-ahead log, in-memory buffering, and SSTable flushes, and adds tagging so data can be filtered naturally. Retention rules help keep storage under control, while a small alerting component highlights unusual patterns or threshold breaches. Overall, ZeitDB aims to offer a clear, approachable system that handles real-time workloads without the complexity of larger TSDBs. This research paper can be useful as a single node to compare different timeseries databases.

Perguntar à IA

Bookmark

View Full Paper