This Log Buffer Edition covers Cloud, Oracle, SQL Server, and PostgreSQL.
Stateful and ML workloads now run better on Google Kubernetes Engine with the latest version 1.9
Three ways to configure robust firewall rules
Why you should pick strong consistency, whenever possible
Simplify Cloud VPC firewall management with service accounts
Announcing our new beta for the AWS Certified Security – Specialty exam
Maximize your VM’s Performance with Accelerated Networking – now generally available for both Windows and Linux
Using Zeppelin to Access MySQL
Dockerizing Your Development and Test Oracle databases
Anyone can be a developer – logic apps from Azure
EBR – Part 5: Explicit Actualization of Dependent Objects
Announcing participants in Oracle Dev Gym SQL Championship 2017
Azure DWH part 23: Data Mining
Secrets to Fast, Easy High Availability for SQL Server in Amazon Web Services
SQL Server 2017 and Python Tutorial
Using MERGE in SQL Server to insert, update and delete at the same time
GDPR – A guide for the perplexed
Thanks to Jeremy Schneider.
Mostly everyone is back into the office now after their holiday vacations, and we’re getting back into the regular rhythm. This week there’s been a healthy pace of articles published in the PostgreSQL community too. Here are the ones that stood out the most to us!
First: I’m more of a data and operations guy, but I really love developers. Why? Because if there were no apps then my data stores would be pointless! PostgreSQL is known for its extensibility, including more available procedural languages than any other engine I know of. This week I saw a blog post that added yet another language to the repertoire for writing database functions (and procedures, starting in PostgreSQL 11). On Jan 7, the Always Right Institute of Germany published a great demo of Helge Heß’ implementation of PL/Swift for PostgreSQL!
Next up, many Oracle DBA’s have learned a lot from Oak Table member UweHesse through his books, blog articles and speaking engagements. This Monday (Jan 8) he published an article about How to cancel SQL statements and disconnect sessions in PostgreSQL, even discovering a useful capability PostgreSQL has that Oracle doesn’t.
Most likely, your very first guess about the best database schema for your application was exactly right? Ha! No way! After going live, did you ever need to add an index, add a foreign key or drop a column? Yes of course. These DDL operations acquire locks that can impact your database more broadly than you’d expect. On Jan 9, Hans-Jürgen Schönig from CyberTec published a great article called ALTER TABLE: High-Availability taken care of. He shows how to change your data structures with absolute minimal impact to your application’s availability.
Moving a little more toward the operations side, there was another great article this week on the ever-important subject of vacuum. On Jan 5, Robert Haas from EnterpriseDB traced the history of vacuum development all the way back to version 8.3 and then summarizes the current state today. This is a must-read for anyone who uses PostgreSQL – even on managed cloud platforms.
Finally, my favorite part: the deep dive into internals. Lets start with an article posted to gitlab on Jan 6: D.S. Ljungmark gave a concrete example where the physical ordering of data on disk in PostgreSQL made an identical query over 10x faster (from 385ms to 27ms). Wow!
And now that you’re so interested in the on-disk data, how to you start looking at it? Luckily, just this Monday (Jan 8), Peter Geoghegan came to our rescue with an article on updates to the impressive pg_hexedit tool! I can’t wait to get my hands on it!
But of course one tool isn’t enough, so lets have two! This second one coming out of that indefatigable Oak Table gang. Back on Dec 21, we somehow forgot to mention Bertrand Drouvot’s article about his tool pgdvf (PostgreSQL data file visualizer), inspired by Frits Hoogland and Kamil Stawiarski. Better late than never! These were our favorite articles this week. Enjoy!