Technical Articles
F5 SMEs share good practice.
cancel
Showing results for 
Search instead for 
Did you mean: 
Lori_MacVittie
F5 Employee
F5 Employee

0151T000003d70vQAA.pngBack in 2008 we noted that when it comes to cloud computing, vertical scalability was still your problem.

Since that time, scalability – in general – has become a MUST have rather than something business hopes to need one day. Thanks to digital transformation and the rapid pace of technical innovation across apps and devices, scalability needs to be “baked in” to each and every application architecture.

So it’s no surprise that cloud remains a significant market force. At the heart of the operational value proposition of cloud lies scalability, the very thing that’s so critical to enabling business growth in today’s digital economy. In fact, scale has risen to dominate the top adoption drivers and benefits of cloud computing models.

What we’ve seen in the past decade is no change in the responsibility for vertical scalability. If you need more CPU or more memory, well, get thee a phatter server, my friend.

But the thing is that horizontal scale – the kind cloud excels at providing – has become the norm for scalability. Emergent trends in application architectures like microservices and APIs have combined with the rise of container-based environments to provide exactly the kind of architectures that are designed to horizontally scale out to heights we could never reach using vertical scalability.

It’s true that there are still limitations to horizontal scalability, particularly in on-premises, private cloud implementations where the ability to re-architect the entire network is not possible. There are still network-hosted components that are best scaled out vertically – phatter pipes, higher port densities, and the like – but we continue to see rapid advances in the ability of cloud “stacks” to enable the same kind of horizontal scale of network components as we enjoy for application components.

So yes, vertical scale is still your problem. The good news is, however, that more and more frequently the need for vertical scale is shrinking as we get better and better at horizontally scaling everything in the data path.

Version history
Last update:
‎23-Jun-2017 04:44
Updated by:
Contributors