Autoscaling

Handle peak loads automatically, without nasty surprises on your final invoice

Is your website traffic as unpredictable as the weather? If so, you could provide the highest resource capacity all year round, but that would be a waste, and you would also be paying unnecessary costs. Would it not be better to have your applications and services scale dynamically, depending on your capacity requirements? Thanks to our autoscaling feature, you can experience as many peaks as you want, while paying only for what you need. Our experts keep a constant eye on this!

Optional feature for cloud users
Also available on Kubernetes for customers who want to work using containers
Automatic peak load management when necessary
Continuous monitoring so that your traffic peak remains manageable

Schedule a meeting

Call us: 09 218 79 79

Available now

Scaling feature adapted to your needs

In an ideal world, you know exactly when your website or application can expect a peak in the number of users. You can then perfectly anticipate this situation thanks to your systems. But what if something unexpected happened, such as a global virus outbreak that would encourage people to go digital en masse? You always want to give your visitors the very best browsing experience, but you are not willing to pay for resources that you are not using. That sounds logical, right?

That is why we at Combell offer an optional automatic scaling feature, completely tailored to your needs. How does it work? Based on an in-depth preliminary interview, we decide which capacity (CPU) and how much memory you need and we set a consumption limit. These criteria help determine the price of your one-time fee. This means that with us, you will not have to deal with unaffordable standard packages; instead, we will provide you with a solution that is completely tailored to your needs.

Controlled scaling, without nasty surprises

The autoscaling feature allows you to add your applications to a pool of resources so that traffic peaks can be handled when necessary. We continuously monitor the load for you and automatically scale the server up to a predetermined limit. If you are about to go over your limit, we will notify you so you can decide if you want to use additional capacity.

If you give us permission to scale up capacity, we will add the required capacity and memory – no more, no less. And without any downtime. This means that at the end of the month, you only pay for what you actually used. Once the peak is over, you can scale down again, in order to avoid unnecessary costs.

Flexibility in both directions, without a fixed contract
Monitoring of your usage and control over your peak usage
Variable limit, under which you stay or that you go above as needed.
No nasty surprises when you receive the final invoice: you pay for your monthly fee plus any additional usage (the cost of the CPU and memory)

Scaling up or down: how do we proceed?

For cloud servers

For our cloud customers, we use the vertical scaling technique, also known as 'scaling up'. We increase the capacity or memory of your system by adding more powerful resources to the virtual machine. This is a fairly simple process, which is therefore quick to complete. Moreover, this technique is cheaper than having permanently higher resources, because you only pay for what you use when traffic is peaking.

On Kubernetes

For Kubernetes customers, we can also scale horizontally (or 'scale out'). We add extra virtual machines, making the whole system more powerful. The big advantage is that you avoid downtime on your servers when downgrading.

Award-winning solutions

Smart Business Awards Hosting & Domain Names 2017

Twinkle Awards Hosting & Domain Names 2016

Autoscaling vs application load testing

If your website or application is working properly, but you want to be prepared for any unexpected increase in usage, autoscaling is just what you need.

With application load testing, you are taking a step back. This test helps you determine whether autoscaling is an interesting feature for you. Or better yet: if problems arise, the test checks whether your application can handle a traffic peak. Because just adding resources will not solve anything if there is a problem with the functionality of your application or website, even under normal load conditions.