Technical Track

Log buffer #546: a carnival of the vanities for DBAs

4 min read

May 23, 2018 12:00:00 AM

This log buffer edition covers Cloud, Oracle and PostgreSQL. Cloud: Google Maps platform now integrated with the GCP Console Getting more value from your Stackdriver logs with structured data Improving application availability with Alias IPs, now with hot standby Performing a large-scale principal component analysis faster using Amazon SageMaker Optimized TensorFlow 1.8 now available in the AWS: deep learning AMIs to accelerate training on Amazon EC2 C5 and P3 instances Oracle: Using GoldenGate LogDump to find bad data Partition-Wise Operations: new features in 12c and 18c SOA Suite 12c in Docker containers: only a couple of commands, no installers, no third party scripts Checking if the current user is logged into Application Builder PostgreSQL: Let's start out with some fun! I really enjoyed Wendy Kuhn's article on May 5 about the history of PostgreSQL. She starts out by relaying the importance of learning the history behind new technical tools & concepts when you're learning. I couldn't agree more. Speaking of history, I've been waiting for the right time to mention this fun article from August 2016. Now is the time, because it relates to the previous article and because I saw a few retweets last week mentioning it. Did you ever wonder why the PostgreSQL logo is an elephant? Or what his name is?? Or even better - did turtles or cheetahs ever represent PostgreSQL???? Patrycja Dybka answers these questions and more. Check it out! Ok, moving on to the serious stuff. :) First off, we've got a new round of minor releases of PostgreSQL. Version 10.4, 9.6.9, etc, were released on May 10. Time to start planning those upgrade cycles! Next up, JD posted a nice summary of PGConf US in New Jersey on May 7. I saw a lot of familiar faces in his pictures! One quick call-out: I heard good things about the speed mentoring at the career fair. I think that was a great idea. (Among many at PGConf.) Another interesting thing JD touched on in his blog post was the growing role of larger companies in the community. He gave a few specific examples related to Google and Microsoft. Back on April 17, Pivotal published an article listing a number of specific ways they contribute to PostgreSQL development, as well. Speaking of cloud companies, who doesn't like a nice rowdy comparison? Over on the SeveralNines blog, we got exactly that on May 1: a quick side-by-side comparison of a few of the many cloud providers who ship PostgreSQL. There are a bunch more - feel free to leave comments on their blog post with the providers they left out! As long as we're doing comparisons, I saw this old website on Twitter last week, and it's fun enough to pass along. Thomas Kellerer from Germany wrote a nice open-source tool called SQL Workbench/J. In the process of supporting many different databases, he's learned a lot about the differences between them. And his website has a really detailed list. Check out this list of SQL features by database - PostgreSQL is looking good! I always enjoy a good story. Singapore-based Ashnik recently published a new case study about a global insurance company who deployed a bank data exchange system on PostgreSQL: a fine example of the serious business that runs on PostgreSQL every day. Moving into the technology space, infrastructure company Datrium has recently published a series of interesting articles about the benchmarking and heavyweight workloads they're throwing at PostgreSQL. The most recent article on April 25 discusses PostgreSQL on bare metal and it has links to many previous articles. In the category of query tuning, how would you like to make a small tweak to your schema and SQL, then experience a 290x speedup? That's exactly what happened to Yulia Oletskaya! She writes about it in this article on May 7. "What's common between DBA and detective? They both solve murder and mystery while trying to make sense of the nonsense." That's the first sentence of Alexey Lesovsky's April 17 article about troubleshooting a PostgreSQL crash. Going a little deeper, I have a handful of recent articles about specific database features in PostgreSQL. First, how about a demonstration of PostgreSQL's top-notch built-in support for full-text search? What better example than analyzing the public email of PostgreSQL contributor Tom Lane to find what his waking hours are? Turns out that he's very consistent. In fact, it turns out you can use Tom Lane's consistent email habits to spot server timezone misconfigurations. Citus also published a nice article back at the beginning of April about row-level security. I didn't include it last month but it's worth mentioning now. PostgreSQL's capabilities here are quite nice. My past newsletters have been following Dimitri Fontaine's series on PostgreSQL data types. We've got three new ones this time around: JSON, Enum and Point types. A big selling point for PostgreSQL is its extensibility. On May 8, Luca Ferrari from Italy published an article in BSD magazine which walked through the process of building a customer extension to provide a new foreign data wrapper that connects the database directly to a file system data source. Our friends at Timescale put out an article about streaming replication on May 3. Lee Hampton gives one of the best descriptions of this critical HA concept that I've seen anywhere. Finally, can a week go by without new articles about vacuum in PostgreSQL? It seems not! On Apr 30, Jorge Torralba published an article on DZone about tuning autovacuum. He has a specific focus on bloat, which is an important reason for vacuuming. There are some nice examples here. And back on April 3, Emily Chang from Datadog published perhaps one of the most detailed articles about vacuum that I've seen. Well worth reviewing. To close up this edition: something a little different. This year marks the 15th anniversary of pgpool. And Tatsuo Ishii reminded us with a blog post on April 15. So in honor of the 15th aniversary, let's collect a few recent links * just* about pgpool! Tatsuo also published two other articles in April about various features of pgpool: And Vladimir Svedov at severalnines published a two-part series on pgpool in April as well. And that's a wrap for this week. Likely more content than you'll have time for, as usual! My job here is done. :) Have a great day and keep learning!

On this page

Ready to unlock value from your data?

With Pythian, you can accomplish your data transformation goals and more.

Speak with Pythian consultants now →

What We Do

Log buffer #546: a carnival of the vanities for DBAs

Share this

Share this

More resources

Advanced compression option caveat in Oracle 12c

How to use createGoldImage For Cloning in 19c

Mystery Solved: Replication lag in InnoDB

Ready to unlock value from your data?