RustProof Labs: blogging for education (logo)
My book Mastering PostGIS and OpenStreetMap is available!

PostgreSQL vs. MySQL: Why we use PostgreSQL

By Ryan Lambert -- Published August 01, 2016

I fully believe that PostgreSQL is the best choice for a reliable and high performance database. This post was inspired after reading a post on Udemy concluding with MySQL being better than PostgreSQL. There are a number of points I disagree with in that post so I decided to make my case of why I think PostgreSQL is a better choice in most situations. In this post I start by countering a few statements from the Udemy post, providing examples from my experience and other resources. Towards the end I go over a few of the PostgreSQL features that I believe makes it truly stand out.

If all you need is to support 3rd party software that supports MySQL (and not PostgreSQL), then you obviously don't have a choice and this discussion is irrelevant. On the other hand, if you're looking for a powerful, reliable, and open source relational database platform:

PostgreSQL is my recommendation.

Continue Reading

Testing PostgreSQL Unlogged Tables for Performance

By Ryan Lambert -- Published July 31, 2016

PostgreSQL has a feature called UNLOGGED tables. I became intrigued by UNLOGGED tables after reading this post. According to the docs:

"Data written to unlogged tables is not written to the write-ahead log (see Chapter 29), which makes them considerably faster than ordinary tables. However, they are not crash-safe: an unlogged table is automatically truncated after a crash or unclean shutdown."

It's important to note that unlogged tables shouldn't be used for storing critical data. My use case for them is to quickly load data to staging tables before cleaning and reorganizing. I end up importing large amounts of data on a regular basis, so this is of interest to me. I decided to test regular tables vs unlogged tables by importing 6 months of NOAA's hourly weather data. You can read about my previous tests with this data here.

Continue Reading

Revisting my Tool Set

By Ryan Lambert -- Published May 03, 2016

It's been over two years since I first wrote about the tools I use regularly as a programmer. Some things have changed, others have persisted. In this post I go through my current tool set with some explanations of why things have either changed or stayed they same.

Some tools haven't changed a bit. Git, SourceTree, VirtualBox and Jenkins are four tools that I still can't live without. Also, my MacBook is now even older, but with a new SSD it seems to have new life.

Continue Reading

Small Community, Big Changes -- The Impact of a Mapping Party

By Ryan Lambert -- Published April 16, 2016

RustProof Labs just completed another Maptastic mapping party! The event was held on April 9, 2016 in Golden, Colorado. Our home base for the event was with the fantastic folks at Pangea Coffee Roasters. Matt, Amanda, and the rest of their friendly staff couldn't have been better hosts for this event! We had 14 total volunteers with 5 of them being "mini-mappers" under the age of 15. My favorite part of the event was seeing the sheer excitement, thrill and enthusiasm of our mini-mappers (and adults) who participated.

Our Volunteers and Pangea Coffee Roasters all deserve a huge Thank You! Without you, events like this are not possible.

The main goal of this post is to visualize the incredible impact our team of volunteers had on the data for our local community. I also want to discuss a few reasons why I think OpenStreetMap is such a valuable project. Before I go on, check out the first time lapse showing how the data in Golden has improved. Feel free to take your time and really watch how the map has improved in various areas.

All data used throughout this post is © OpenStreetMap Contributors.

Continue Reading

Stop Using SSL, Use TLS Instead

By Ryan Lambert -- Published March 24, 2016

Security matters. One of the bigger security issues I see in our online security that too many people are supporting outdated, insecure technologies. One of the big offenders is "SSL" (Secure Socket Layer). Recently, a new attack against SSL v2 was discovered, dubbed the "DROWN" attack. Part of what makes me wonder about this, is why do systems in 2016 still support protocols that have been known to be insecure for more than 5 years? Further, even the most recent version of the SSL protocol (v3) has been insecure since 2014 when the POODLE vulnerability was discovered.

In reality, SSL has been antiquated for quite some time being superseded by "TLS" (Transport Layer Security).

SSL Is Insecure; use TLS instead. I want to make clear that there is no secure SSL protocol remaining.

Continue Reading

<-- Older Posts          Newer Posts -->