Home

Welcome to the cockroach wiki!

Roadmap

CockroachDB is currently in beta. We track issues using 3 milestones:

• Q2 is for work to be completed in the current quarter.
• 1.0 is for work to be completed by the 1.0 release.
• Later is for work to be completed after the 1.0 release.

Issues without a milestone have either not been scheduled or are cleanup work that can be done whenever someone finds the time or inclination.

The goals for Q2 are increasing stability and performance at scale, fleshing out additional SQL functionality and kicking off work on distributing SQL execution.

• Core & Stability

  • Continuous load testing on a 10 node cluster
  • Test clusters with increasingly large range counts (10k ranges, ~500GB)
  • Improve rebalancing heuristics
  • Verify on-the-wire and on-disk compatibility between versions
  • Infrastructure for data migrations

• SQL Functionality

  • FOREIGN KEY constraints
  • CREATE SEQUENCE / SERIAL column type
  • Dump/restore of SQL data+metadata
  • Improve type inference
  • Improve testing of postgres-client compatibility

• Distributed SQL

  • Push-down filtering to ranges
  • Push-down updates to ranges
  • Infrastructure for distributed joins / aggregations
    • Stretch goal: inner joins

• UI

  • Adjustable graph timescales
  • Flesh out databases and tables pages
  • Tooltips on all graphs
  • Provide standard hooks for timeseries data (e.g. Prometheus)

Frequently Asked Questions (FAQ)

Why the name Cockroach?

It is evocative of the properties of the system we are trying to build. Cockroaches are hard to kill, spread easily, and evolve quickly. “Roach Motel” would also be a great name for a hosting service.