# How Discord Automates ScyllaDB Clusters at Scale

DevFeed: [How Discord Automates ScyllaDB Clusters at Scale](<https://devfeed.tech/articles/how-discord-automates-scylladb-clusters-at-scale-246.md>)

Original publisher: [Read original article](<https://discord.com/blog/how-discord-automates-scylladb-clusters-at-scale>)

Author: Peter French

Published: 2026-05-08T00:00:00Z

Content type: article

Language: en

Sources: [Discord Blog](<https://devfeed.tech/sources/discord-blog.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [postgresql clusters](<https://devfeed.tech/topics/postgresql-clusters.md>), [Replication](<https://devfeed.tech/topics/replication.md>), [Concurrency](<https://devfeed.tech/topics/concurrency.md>), [configuration](<https://devfeed.tech/topics/configuration.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [Rust](<https://devfeed.tech/topics/rust.md>), [YAML](<https://devfeed.tech/topics/yaml.md>), [Prometheus](<https://devfeed.tech/topics/prometheus.md>), [API](<https://devfeed.tech/topics/api.md>)

Tags: [concurrency](<https://devfeed.tech/tags/concurrency.md>), [database](<https://devfeed.tech/tags/database.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [prometheus](<https://devfeed.tech/tags/prometheus.md>), [replication](<https://devfeed.tech/tags/replication.md>), [rust](<https://devfeed.tech/tags/rust.md>), [scylladb](<https://devfeed.tech/tags/scylladb.md>), [workflow](<https://devfeed.tech/tags/workflow.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

## AI overview

Discord describes how it automated ScyllaDB cluster operations with a Scylla Control Plane built around tasks, workflows, and jobs. The system uses explicit condition checks against Scylla's API and Prometheus metrics, idempotent Rust tasks, and YAML-defined workflows with retries, parallelism, configuration, and runtime templating.

## Source excerpt

You've been asked to stand up a brand-new database cluster, meaning a whole day of configuring dozens of nodes, validating replication, wiring up dual-write pipelines... what if this whole ordeal took less than two hours? This is how we got ourselves into this exact mess, and how we made our way out of it.