Extinguish the routing tire fire.

How infrastructure evolves, and how Backplane can help.

In our last post, we introduced one of the key technologies that makes it possible for Backplane to help you untangle the network that runs your web architecture. This time, we’re going to walk you through how a typical web architecture evolves, and how you can avoid a whole lot of pain by using Backplane instead.

Sign up now to get started for free.

How a web architecture evolves

All web architectures start out the same way — with a whole lot of good intentions. As business needs take over, time becomes an issue, that round of funding sneaks up on you, etc., a lot can go wrong. Here’s how it usually goes:

1. We need a web service! I know, we’ll just launch an instance in AWS and set it to listen on port 80. Boring is good, right?
2. Oh well, it went down once and we realized we probably need some redundancy. Okay, let’s ship another instance, and stick them behind a load balancer. Ahh, good. All better.
3. Shoot, there’s some DNS issues involving communication with an external server. How do we handle that? Update DNS when one goes down? Okay, let’s write some scripts.
4. Turns out that the load balancer needs some love too. We’re writing more infrastructure code than we wanted to. I guess we need some tests for those? More code equals more bugs, we know that much.
5. Okay, things are growing and we need more instances. How are we going to get everything consistent? Service discovery? How much consistency do we really need? Should we use internal DNS?
6. We want to roll traffic over per-request instead of per-connection, how do we do that? What proxy plugins do I have to compile and mash together to make that happen?
7. What about detailed logging and metrics? :/
8. Time to configure and maintain that firewall. Who can get in again? How do we define safe traffic routes within the Data Center? How do we talk between Data Centers for that one enterprise customer?
9. Wow, that security compliance questionnaire was considerably more involved than I thought it was. We should audit all of our TLS certs. How do we manage them? Can’t someone do this for me?
10. How can we coordinate health checks? How do I know when things are running? How much more infrastructure code do I have to write before I can go back to working on the product?

Use Backplane instead

Backplane is designed to be your reliable, consistent interface to help you manage your infrastructure. Get health checks, intelligent / secure routing with end-to-end TLS, mutual authentication between data centers, load balancing, and a growing list of advanced middleware like: OAuth flow handling, geo headers, ip rules, and more! All handled close to your users without having to write a single line of code. Backplane requires no extra infrastructure, is secure by design, and is trusted by experts.

Where you used to have a fractured set of services with different configurations and control planes, you now have one interface to help you untangle your network: Backplane. Less code means fewer bugs, and fewer tire fires.

Let us help you untangle your network — click here to sign up!