https://github.com/jdockerty/lynx
A toy time series database, built atop of Apache DataFusion.
https://github.com/jdockerty/lynx
database rust time-series
Last synced: about 2 months ago
JSON representation
A toy time series database, built atop of Apache DataFusion.
- Host: GitHub
- URL: https://github.com/jdockerty/lynx
- Owner: jdockerty
- Created: 2024-12-18T20:57:38.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-27T21:22:34.000Z (over 1 year ago)
- Last Synced: 2025-02-28T06:16:27.902Z (over 1 year ago)
- Topics: database, rust, time-series
- Language: Rust
- Homepage:
- Size: 116 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 8
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# lynx
An in-memory, time series database. Built on top of Apache Arrow & DataFusion.
>[!WARNING]
>This project is not intended for production use cases.
## Overview
At a high-level, measurements are ingested through the HTTP API and made durable in the write-ahead log (WAL) before becoming queryable within the in-memory hierarchical buffer. The WAL also ensures that crashes can be tolerated before data has made its way into the buffer.
Ingested data is stored hierarchically through the combination of a namespace and table. Within a table, data is partitioned by time although this is currently fixed to daily partitioning and is not configurable.
Executing a query means that in-memory data is converted into Arrow before making its way through DataFusion for processing and returning to the client.