Introduction

Dashiki is intended as an at-a-glance dashboard for time-series data from multiple sources, allowing operations, engineering or business to view a handful of related metrics at a time. It is possible to style metrics with threshold function to flag possible issues, and discrete events (deploys, migrations, backups) can be overlaid to expose correlation of incidents.

It is not intended as a comprehensive data exploration tool, instead preferring to provide links to the original data source where possible. Though originally written as a graphite front-end (and that bias is probably still apparent), the data retrieval code is modular and it is hopefully trivial to write plugins for any service that exposes time-series data over HTTP.