GatewayAPI is a complex system that makes sure that your SMS-traffic is sent quickly and securely to places all over the world while making sure that the prices are low and the system is reliable. That is by, all means, no small feat.
In this blogpost we will touch upon how we handle daily operations, monitoring as well as security. We will also reveal what we are currently working on to further heighten the standards of GatewayAPI.
Here we will look at how our datacenter is set up, how we monitor and what different processes we have implemented to ensure that everything keeps running smoothly.
Tier 4 datacenter setup (based on European standards). The definition of a Tier 4 data center is a location with high security and redundant and dual-powered servers, storage and network links. Components are powered with multiple and independent sources of power and cooling resources. All this guarantees a high availability. In the last 90 days (from april 9) our uptime on the API has been 100 %. (Go to https://status.gatewayapi.com to keep tabs on the status). Backup of our servers are made daily and it is possible to restore any prior version of the system dating back six months
We handle our monitoring with Icinga and Nagios. Additionally, our office has been converted to a NASA-inspired command center with 3 screens feeding our team continuously with relevant updates such as SMS speed, queues, deliverability, operators, datacenters etc.
4-eyes principle: Every action has to be approved by two members of our staff before it can be implemented. This improves decision making and helps eliminate errors. We also collaborate to make sure that all our solutions are updated to the newest versions.
Redundant operators: Messages are always routed through the best suited mobile network operator, based on message class, destination, traffic load etc. If one of the operators run into downtime etc., we can quickly switch between them.
What we are working on
We are currently working on deploying in a new datacenter to further increase the guaranteed uptime.
We are also working on containerizing everything and deploying it on Kubernetes which builds upon 15 years of experience of running production workloads at Google. Kubernetes defines a set of building blocks (“primitives”) collectively, which together provide mechanisms for deploying, maintaining, and scaling applications. Essentially we will be able to scale GatewayAPI effortlessly when the deployment has been completed and it will further improve the way we handle daily operations as well.
That was it for this time, folks!
If you have any further questions concerning our operations, monitoring and security don’t hesitate to contact our support team.
Subscribe via RSS