Modern cloud services must be able to handle user load dynamically and flexibly.
Vertical scaling is not an option when building modern cloud services. It might work for a while, but inevitably, one of two things will be true; either you'll be overprovisioning (and therefore overpaying), or you'll be vulnerable to maxing out your capacity.
But the good news is that horizontal scaling can be implemented with relative ease in today's day and age, and that's what we do with every service that we build, whether self-hosted by our clients or managed and hosted by us on behalf of our clients.
Everything we design and build is fully redundant and scales automatically as load increases; this includes networking infrastructure, compute infrastructure and both storage capacity and compute infrastructure on the database platform. In addition, all system components are designed to remain fully functional and retain their highly available qualities during the scale-out and scale-in process.
Getting the most out of your Database with Intelligent Query Distribution
The cloud services that we design and build have complete control over how they utilise the data platform that serves them.
This enables the system to target secondary nodes within the data tier with read queries for resources that are not fast-moving or which do not have strong data consistency requirements. Reducing the load on the primary node within the data tier has significant advantages to the system's overall performance and cost-efficiency.
While this may sound complicated, the good news is that it's all handled by our framework, meaning no code has to be written to gain this incredible advantage.