Introduction
A scalable system can adapt itself by providing more or less resources to meet demands. Nowadays, with the unpredictability of monoliths and microservices systems, it’s important to always build a system to be scalable (or to be ready to) no matter if we are dealing with computing, storage, database, and any other components.
The YouTube Channels in both English (En) and French (Fr) are now accessible, feel free to subscribe by clicking here.
Scalability Types
There’s no single way to make a system scale, sometimes it’s done by offloading some tasks to other threads, but in this section, we’ll focus on the main Scalability Types applied to IT systems, which are the vertical and the horizontal approaches, respectively referred to as scaling UP and scaling OUT.
Horizontal Scaling
Working on scaling your system horizontally means making sure that more compute resources, generally with the same specs, are added to your infrastructure. To make it simple, if one computer (VM: virtual machine) was used to process your web App requests, with more processing power needed, more computers would be added to the float while making sure to use a Load Balancer that distributes the workload amongst them.
Despite it being complex to implement, one of the biggest advantages of Horizontal Scaling is the increased redundancy since the App or one of its components is getting duplicated into a completely different device to support more traffic, which means, that even if one of the duplicated parts goes down, the system won’t fall so it’s good to avoid single point of failure.
Vertical Scaling
The purpose of scaling remains, but the approach differs, with Vertical Scaling, instead of adding more computers to our infrastructure, we’ll increase the specs of the existing one. To be more specific, an example could be to move from a virtual server of 32 GB of RAM to 64 GB. Any removable spec as the CPU could be into play, not only the RAM.
Contrary to the precedent, it’s easier to apply this approach and to manage vertically scaled systems. But if it’s the only method that our system benefits from, we should watch it carefully since there’s a big risk of downtimes.
Mixed Scaling
In reality, depending on the system, it’s possible to mix up both scalability types in order to reach our goals, putting together the vertical and horizontal ways for our App to be as responsive as possible according to our target.
Scaling Policies
There are many approaches for the scaling activities to be launched so that our infrastructure changes to accommodate the traffic, the most common are:
- Manual scaling: in which you’ll personally handle the scaling by manually adding and removing instance(s) to the pool.
- Dynamic scaling: where some conditions are to be met for a “scale activity” to launch automatically, they can be of many natures depending on what you are measuring your infrastructure operations with. A system could scale based on the usage of the following metrics: the CPU, Latency, Memory (RAM), and so on.
- Scheduled scaling: scaling based on a schedule (date + time) gives you the possibility to scale your infrastructure in response to predictable traffic. For example during Christmas or Black Friday.
Cloud Scaling Services
Cloud enables us to accomplish even more quickly, in case you opted for moving into the Cloud, depending on the public Cloud provider that you use, some services are available to scale infrastructures:
- AWS Auto Scaling Group
- Azure Scale Set
- GCP Managed Instance Groups
Let’s note that the Elasticity property of the Cloud allows your system, not only to scale UP and OUT but also to scale DOWN and IN to release unused resources once the traffic decreases.
If you are interested in a demo of how we scaled OUT/IN a project migrated to AWS public Cloud, the following video is for you:
———————
We have just started our journey to build a network of professionals to grow our free knowledge-sharing community that’ll give you a chance to learn interesting things about topics like cloud computing, software development, and software architectures while keeping the door open to more opportunities.
Does this speak to you? If YES, feel free to Join our Discord Server to stay in touch with the community and be part of independently organized events.
———————
Conclusion
This article highlighted the different types of Scalability Types, and their particularities as well as pointing out what the possibilities are in terms of services while using public Clouds like Amazon Web Services, Microsoft Azure, or Google Cloud Platform.
Thanks for reading this article. Like, recommend, and share if you enjoyed it. Follow us on Facebook, Twitter, and LinkedIn for more content.
