How Discord Indexes Billions of Messages — Discord is a chat service supporting millions of users. How could they add quick searching? Here’s how they set it up with Elasticsearch.

Timescale: An Open Source Time-Series Database — SQL made scalable for time-series data. It’s Postgres compatible and optimized for fast ingest & complex queries.

Real-Time Deduping At Scale with Redis — Deduping at scale is a hard problem and Tapjoy share a lot of details of how they use Redis to do it on over 2 million analytics messages per minute arriving in an ‘at-least-once’ style.

Linode is the SPEEDIEST SSD host for your DB - simple, reliable & powerful. — MySQL? MariaDB? PostgreSQL? Whatever DB you use, run it on a Linode, the most scalable, reliable and fastest servers in the cloud. Use promo code DB20 for a $20 credit and get started!

Miller: Like awk, sed, cut, join, and sort for CSV, TSV, and JSON — Provides Unix tool-esque functionality but around key-value pair data. Written in C too, so no dependencies.

Hoodie: Uber's Incremental Processing Framework on Hadoop — Uber Engineering recently built and open sourced Hoodie, an incremental processing framework to power Uber’s data pipelines at low latency.

Has Hadoop Failed Us? — Alex Woodie brings together several people from across the industry to question if the Hadoop dream of unifying data and compute in a distributed manner has become too expensive and complex.

Don’t Migrate Databases Automatically — Migrating your DB and deploying app code at the same time can create a race condition that could take down your app.