Fastly and DataDog

A link to this was posted in the TMF premium boards, but Fastly recently release a video on how they work with DataDog:

There is an engineer from both companies in the video. The first 10 minutes or so is a good introduction to both companies.

For Fastly, I knew about the potential for load balancing, but I hadn’t realized that since Fastly handles not just static but dynamic content, it can completely off-load your “origin servers” (the ones that have the data), and as such lets you do things like migrate from one cloud provider (AWS, Azure, GCP, etc.) to another without your users ever being aware. You can even stitch content on multiple clouds together seamlessly. And, of course, since all your user/comsuming traffic is going through Fastly, you can both keep high security on your origin servers (since the only thing they send to is Fastly), as well as rely on Fastly to scale up and down as demand shifts for you.

For Datadog, if you haven’t seen the product in action, there’s a live demo starting about 16 minutes in.

Most of the video is about how the two companies’ products integrate. DataDog has lots of integrations, of course. The integration with Fastly is interesting since one of Fastly’s attributes is that it generates and sends lots of log information in real-time. So, you pipe those into DataDog and can gain instant insights into what’s happening with your users. They give examples of real-time debugging of issues and such. It’s quite long and even I couldn’t stick with it.

At any rate, it’s good seeing this from the engineers’ perspective. These companies both have great products.