#214 — July 27, 2018

Read on the Web

Database Weekly

Machine Learning in Google BigQueryBigQuery ML extends BigQuery with the ability to build machine learning models on massive structured or semi-structured datasets using the same SQL data analysts/developers are familiar with.


An Introduction to Elasticsearch SQL with Practical Examples — Released just over a month ago, Elasticsearch 6.3 added initial SQL support to the popular search-oriented document database. Here we see how that works in practice.

Dale McDiarmid

5 Key Decisions to Make When Moving Oracle Workloads to the Cloud — This EDB Tech Guide covers five key decisions you’ll need to make to navigate your choices and enable your success.

EnterpriseDB sponsor

The Evolution of Application Data Caching at Netflix: From RAM to SSD — Netflix was storing petabytes of data in a memcached-based cache but using SSDs is significantly cutting costs while still getting the work done.

Netflix Technology Blog

The Best Database-as-a-Service Solutions of 2018 — A high level comparison of major database service providers including Microsoft Azure, MongoDB Atlas, AWS, Google, IBM and SAP.

Pam Baker (PC Magazine)

PDF: How Postgres's SQL Dialect Stays Ahead of Its Competitors — Yes, it’s a slide deck, but it’s a Markus Winand slide deck and full of examples. I found a possibly unofficial video of the associated talk if you’re curious.

Markus Winand

Tuning InnoDB Primary Keys — The choice of good InnoDB primary keys is a critical performance tuning decision on MySQL.

Yves Trudeau (Percona)

Introduction to Failover for MySQL Replication


MongoDB 4.0 Is Now Generally Available. Try It Now on MongoDB Atlas — It’s now even easier to address a complete range of use cases with MongoDB.

mongodb sponsor

An Introduction to Azure Cosmos DB Emulator for Building Apps Locally — Azure Cosmos DB is Microsoft’s multi-model cloud database platform but you can set up a local emulator to make your app development process easier without being connected to Azure.

Suhas Pande

UUIDs and Compressibility — Universally unique identifiers, or UUIDs, are often used for database primary keys in scenarios where coordination of persistence is either impossible or impractical. But.. “using UUIDs instead of naturally occurring composite keys introduces randomness where there could often be order.”

Richard Startin

▶  Building a Petabyte Scale Warehouse in BigQuery

Google Cloud Next 18

18 Places, Books, and Pages to Learn SQL — A variety of sources for online training and tutorials to up your SQL skills.