Citus Newsletter

Building real-time analytics dashboards with Postgres and Citus

One use case we’re seeing a lot these days: using Citus to power customer-facing, real-time analytics dashboards, even when dealing with billions of events per day. So here’s some fun reading from Craig: an end-to-end guide to building real-time analytics dashboards, from ingesting large data, to in-database rollups, to massively parallel querying.
Guide to building analytics dashboards → 

Citus Cloud Retrospective for 12/14/2017

On Thursday December 14th, we experienced our first system-wide service outage across Citus Cloud since launching our cloud database service in April of 2016. This retrospective walks through the incident timeline, our learnings, and the changes we’re making to reduce the likelihood of this happening again.
Read the retrospective → 

Distributed count distinct vs. HyperLogLog

With Citus 7.1, there are several ways to do distributed counts, including perfectly accurate distincts which do not include your partition column, as well as probabilistic sketch algorithms like HyperLogLog. This post compares `count distinct` with HyperLogLog. Which should you use?
Count distinct or HLL?  →

PGConf EU: HyperLogLog, Eclipse, and Distributed Postgres

Back in October, a bevy of Citans flew to Poland to mix and mingle at the largest Postgres conference in Europe. Three of our engineers gave talks at the conference, on topics like HyperLogLog, hacking PostgreSQL with Eclipse, and distributed computing on PostgreSQL. Missed the talks from Burak Yucesoy, Metin Doslu, and Marco Slot, but want to see the presentations? We share the slides here.
Slides from Citus talks at PGConf EU  → 

Citus 7.1: Window functions, distinct, distributed transactions, more

All the things about the stealth release of Citus 7.1, including distributed transactions, the zero-downtime shard rebalancer, cross-tenant support for Windows functions, COUNT(DISTINCT) improvements, SQL coverage for subqueries with reference tables, and more.
Read the definitive post on Citus 7.1  → 

Citus warp: Database migrations without the pain

We rolled out a new database migration feature for our cloud database as part of our Citus Cloud 2 launch in November. Warp migration in Citus aims to make database migrations a whole lot less painful. This post walks through how warp works, and what happens when you warp your database from RDS into Citus Cloud.
Read Lukas’s post on warp migration  → 

Revamped & redesigned Product Overview page for Citus

To make it easier for developers to understand what Citus is and why they should care, we revamped and redesigned our Product Overview page (as well as the product detail pages for our open source, enterprise, and database as a service offerings.) Check it out—and help spread the word if you’re a fan of Citus.
Check out the new Citus product overview  → 

ICYMI: How Citus Executes Distributed Transactions on Postgres

Transactional semantics and ACID properties are already hard to provide on a single node. So how do you provide them in a distributed database, in the face of machine and network failures? Our CTO Ozgun Erdogan explains how Citus does it.
Read Ozgun’s post on distributed transactions  → 

Things People Wrote About Citus (not Citrus)

• InfoWorld: Why old-school Postgres is so hip again

• The New Stack: Citus Data Turns Postgres into a Scalable Distributed Database System

• Startup Grind: What Happens When Your SaaS App Takes Off?

• Speaker Deck: Our CTO’s slides from his awesome QCon San Francisco talk

Is Citus right for you?

Want to explore whether Citus is right for you and your use case? Learn more about the Citus product, check out the documentation, or let us know if you want to book a demo.