Today, we're announcing critical new observability features for Compute@Edge — logging, tracing capabilities, and granular, real-time metrics. These features, which are now available in beta, bring observability to the forefront of the serverless compute environment.
Observability has always been difficult, and, as web architectures become more complex, observability can feel downright impossible. In multi-vendor architectures, we often end up trading performance for simplicity, using the lowest common denominator feature set that ensures an application will run on multiple platforms and provide basic operational insight.
When Artur founded Fastly in 2011, it was notoriously difficult for developers to get visibility into the traffic running across their CDN. But for developers to really adopt infrastructure and tools as their own, they need to see inside them. That's why we've spent the last nine years building observability into our delivery and security products by default. And why we're doing it with Compute@Edge, too. Here's how it works:
Logging: Like other Fastly services, Compute@Edge lets you send logs to an endpoint of your choosing. In addition, it provides real-time logs sent directly to your CLI without any setup required. When you’re developing, a `println` on Compute@Edge works the same as a `println` on your laptop. With log data visibility, you can determine the root cause of a host of issues — whether they originate in your infrastructure or your end users.
Metrics: Compute@Edge provides more insight and metrics around your functions — like CPU and RAM use — displaying how the code you have in Fastly interfaces with your origin. Plus, you get support for historical metrics or stats, as well as real-time metrics to better monitor application performance and react in real time for less down time.
Tracing: Compute@Edge honors the tracing parameters that come into our platform and maintain them when they leave our platform, so you can correlate and stitch together the lifetime of a request in a third-party visualization system, like Splunk or Datadog, after for further analysis. It’s easy to set up and lets you trace function performance after deployment.
With Compute@Edge, our aim is to design not just another tool for your toolbox, but one that outweighs its added complexity by solving for multiple challenges with serverless: performance, security, developer ergonomics, and, now, observability. Though we're excited about today's release, we’re really just getting started.
We're working closely with the participants in our beta program to identify additional friction points, features, and use cases around observability. This is just the beginning of our plans for making edge applications faster, healthier, and easier to develop and operate — because Fastly succeeds when developers' work gets easier.