It’s all about the app, but do you have the infrastructure to support it?

It’s all about the app, but do you have the infrastructure to support it?

Anthea Nadin, Aptronics Data Centre Solution Specialist

Whether they improve your employees’ ability to do their job, extend your reach, enable new business services or enhance customer experience, apps offer an efficient, cost-effective method for driving business outcomes, or at least they should. Unfortunately, what’s often misunderstood is the crucial role infrastructure plays in the development and support of those apps. Anthea Nadin, Aptronics Data Centre Solution Specialist, discusses this further to enable you to make the best possible decision.

Why is infrastructure so important anyway?

Infrastructure – compute, storage, networking and security – provides apps with an ecosystem in which they can exist. A poor ecosystem affects the health of the app, hindering its ability to meet its intended purpose. Ensuring the best possible infrastructure means aligning it with the same business goals as you would have when creating the app.

Should an app’s audience be entirely employee-based, or reliant on legacy equipment or specialised onsite hardware, an on-premise data centre will suit you perfectly. But if you require a low latency, high bandwidth connection for South African customers, then you might want to consider a local cloud solution, or alternatively, an international one for a primarily international customer base.

With regard to security, infrastructure needs to be built to accommodate an app’s interaction with the outside world – detection of weaponised attachments or malicious URLs in the case of email, or protection against SQL injection for database-centric apps, for example.

Those are some of the instances that demonstrate the importance of utilising the correct infrastructure. Failing to do so can cause instability, performance degradation and, in the case of customer-facing apps, can damage the brand and a company’s reputation.

Infrastructure should be a consideration from the get-go

The app’s requirements are a solid guideline for the infrastructure required, so asking the team developing the app should be the first port of call. Consider an app’s compute and storage requirements, because whereas some databases could require relatively little storage, they’re often very memory and processor intensive. On the other hand, an email app would require a lot more storage, especially if data archiving and backup is important, and it should be.

Ultimately, the teams responsible for app and infrastructural development should work as a cohesive unit. The wrong infrastructure might not just be unfit for purpose – it can be unnecessarily expensive, and prohibitively so. It’s important, then, that infrastructure be considered from day one, particularly security, because a breach is a potential nightmare for any business.

There’s a right way and a wrong way to handle scalability

A common mistake made by businesses is the over or underestimation of the adoption rate of an app. The result is either poor performance or networking issues in the case of insufficient capacity or excessive cost, while an underutilisation of infrastructure occurs when the capacity far exceeds adoption rates. But that’s easily remedied with the right infrastructure, which, if an app has been properly implemented, should allow you to spin up or down capacity with relative ease. That’s why the more egregious error is the use of a rigid infrastructure, leaving little room for scalability.

If an app’s audience is expectedly small, on-premise infrastructure can work. But cloud’s promise of offloading CAPEX to a service provider is hard to beat for anything else, allowing you near-infinite scalability with little effort. In fact, any service provider that cannot provide additional processors, memory or bandwidth at a moment’s notice, for example, is behind the times.

These are the promises of hyper-convergence and composable infrastructure, a software-centric approach to infrastructure that packages all the resources needed to run a single app or an entire stack into virtualised workloads in a data centre. It transforms infrastructure into a dynamic, mouldable resource, simplifying scalability and extending the range of an app to any device with an Internet connection while improving agility and reducing cost.

Browse our latest issue

Intelligent CIO Africa

View Magazine Archive