Tale of the Whale

There are many reasons to like Trino, formerly known as PrestoSQL. It’s a distributed query engine that scales as big as you’ll ever need, and has tons of connectors for existing data sources. For Resurface, our Pilot and Orca editions are single-container installations, with relatively small footprints — and Trino still gives blazing performance when properly tuned to run inside these small containers.

I’m a big believer in data-driven decisions, but the reality is sometimes we’re less data-driven than we’d like.

A focus on uptime and system health doesn't tell you what you most need to know about your APIs. But you may find out the hard way — when your customers stop buying, and you can't quickly figure out why.

You might expect to need a database administrator to manage all the log data that Resurface stores. But the answer is no -- the Resurface database is self managing, so you won't need a crew of lumberjacks.

Simply put, because third-party data sharing has gotten way out of hand. Customers aren't just paying for their data, they are paying with their data. But with a first-party solution like Resurface, you're back in control.

GraphQL adopters are quickly realizing that APM and logging tools built for REST-style APIs are not a great fit. But for us, native GraphQL support is an obvious next step.

Java, NodeJS, Python and Ruby loggers now all implement the v2 logging API. This makes loggers more flexible and efficient, and in most cases only requires minor changes to upgrade.

The design for our 2.x logging API has been in the works for the last few months. Our NodeJS logger is the first to reach this milestone.

Logging rules can be loaded from local files, using relative or absolute file paths, making rules easier to share between apps.

Loggers now capture host names, so request and response data can be pivoted by host

Logging real usage of Django-based APIs just got a whole lot easier. Same capabilities as our other loggers, but written in pure Python3.