Data Feed #2

Learning

Fantastic content all around from both days of the ScyllaDB Summit: Day 1, Day 2

Distributed transactions: What, why, and how to build a distributed transactional application.

Replication

• Improve logical replication performance in Amazon Aurora PostgreSQL with the new write-through cache.

• Dolt binlog replication from a MySQL or MariaDB.

Tips

• Working with JSON data, using monitoring, and connecting to Dremio in SingleStoreDB.¹

• Data modelling and queries considerations at RavenDB.²

• Guide to sharding and partitioning best practices in CrateDB.

¹ Developer Quick Tips

² Recording: RavenDB Usage Patterns

Hacking

As always fantastic and informative video from Andrey Borodin about detailed buffers statistics in EXPLAIN BUFFERS:

Deep Dive

Insightful analysis of implementing comparing data sampling implementation in DB versus on the client.³

³ Downsampling in the Database: How Data Locality Can Improve Data Analysis

Article from Alexey Milovidov himself, that stretches over quite a long period of time and describes ClickHouse journey to the data obfuscation.⁴

⁴ Five Methods For Database Obfuscation

Short a sweet deep dive into Postgres Write Ahead Log (WAL) files and Log Sequence numbers (LSN).⁵

⁵ Postgres WAL Files and Sequence Numbers

Business

Real-money gaming company saves $1.5 million per regional expansion with CockroachDB.

A very interesting case from a company producing batteryless sensors, that continuously stream data used in all kind of manufactoring industries (chemical processing, district energy, rotating equipment – motors, pumps, fans, compressors, etc.)⁶

⁶ How Everactive Powers a Dense Sensor Network With Virtually No Power at All

Influx year in review 2022.