Sending in the Green Berets

blog-author-qual-it.jpg

By Qual IT | 25 October 2013

In information technology (IT), as in war, sometimes you need a team with special skills and a different perspective to change a risky situation.

In wartime (and those old John Wayne movies), the famous US Army’s Special Forces group, the Green Berets, are brought in to apply their special skills to prevent or resolve a crisis. They have five primary missions: foreign internal defence, special reconnaissance, direct action, counter-terrorism and their original and most important mission: unconventional warfare.

In the IT world, the people with the skills for ‘unconventional warfare’ are infrastructure testers. They are the people with general skills and tools that can take an overview look at the quality of outcomes an IT project will deliver. 

Risk and complexity

Most of those involved in the New Zealand IT sector would accept that the risk appetite around IT is diminishing, as IT projects are becoming larger, more complex and involving more interdependent systems and technologies.

Well publicised problems with major local IT projects, from failing to deliver on objectives through to significant security issues, have made Boards and senior managers more nervous about IT and more focussed on reducing the possibility to delay or disaster.

Projects in the news are just the very visible tip of an iceberg of IT projects that will be failing or at least struggling at any time. Figures from the US put the partial (i.e. not fully delivering on objectives) or complete failure of IT projects in that market at around two thirds of all initiatives.

The direct costs in terms of wasted investment, time and materials; and the indirect costs in terms of lost productivity, the opportunity cost of investing in other productivity improvements is enormous. 

Keeping the water in the balloon

In the age of the mainframe computer the implementation challenges were far less, with a single, integrated system of software and hardware, often from a single vendor over a dedicated network.

Modern IT is far more sophisticated, powerful and ultimately cost-effective, but the challenge of managing the implementation of a large-scale project can be like trying to squeeze a water filled balloon in your hand. As soon as you have one part under control, an issue can emerge somewhere else.

A typical IT project is likely to have multiple integrated software technologies (operating system, database, messaging/middleware, desktop applications, online or smartphone apps) hosted and being accessed by multiple hardware devices (on-premise servers, data centres, networked devices, tablets, smartphones) and involve multiple vendors.

Each component can work well, and to the vendor’s specifications, but it’s at the seams the problems can occur. With any reputable product, the vendor will have quality-assured it, but they can’t test it against everything it will possibly connect with or be accessed by. And the more bespoke elements in a system, the harder it becomes.

Trends like bring your own device (BYOD) are adding another level of complexity to the task, particularly around security implications, and the inability of helpdesks to realistically provide effective support for multiple devices.

It’s in the midst of this complexity that the Green Berets of IT, the infrastructure testers, can offer some value.

What is infrastructure testing?

In some ways infrastructure testing defies definition. There is not even a formal definition in the standards.

That’s because infrastructure testing uses skills and techniques from across the testing discipline, and because it typically focusses on the ‘plumbing’ of IT systems, not always that visible to system end users.

Infrastructure testing projects can involve anything from checking operating system upgrades, server migration and virtualisation, firewall upgrades, network exchange upgrades, user provisioning,  disaster recovery, failover solutions and measuring bandwidth to name a few.

Infrastructure testing needs a broad skillset because basically it requires evaluating the puzzle of different pieces that are put together in a typical IT project. An infrastructure tester needs to understand multiple types of technologies and how they fit together, and often come from outside the traditional role of a functional tester.

An example: the OS challenge

Operating systems (OS) upgrades are a classic example of where infrastructure testing is applied, not always with the desired effect. These projects are challenging because they only happen every 3-5 years, and while largely invisible to end-users, can potentially impact hundreds of important software applications within a large organisation.

Where infrastructure testers can add value is reviewing the design of an OS upgrade, evaluating whether a vendor’s proposed approach will deliver on your business objectives and recommending changes. ‘Out of the box’ a new OS will work well and a vendor will be proficient at implementing it, but it is where specific requirements of your business need to be taken into account that problems can arise.

For example, an OS upgrade might enable more mobile working by key staff, but also raises the risk of sensitive information being exposed and the need for file encryption. A vendor may be expert at implementing the core OS upgrade, but not necessarily focussed on these sorts of broader implications.

Another aspect of OS upgrades where infrastructure testers add value is the ability to rationalise applications. Large organisations can have hundreds of different applications running on their OS, and an upgrade could offer opportunities to move from 25-30 versions of one product to a single, certified and properly supported version.

Effectively applying an infrastructure testing team in a project like an OS upgrade may mean only 20-30% of their effort is spent on actual system testing. They are of much more value if they are spending significant time on analysis and design, as well as testing applications and integration.

Data from Qual IT infrastructure projects shows large organisations can potentially save months of project time in an OS upgrade by undertaking sufficient up-front analysis and design.

The green berets

Any army has specific skills and technologies to make them more effective - sharp shooters, infantrymen, remote drones and so on, but sometimes there is a place for people with a broader view and a range of skills – the Green Berets in military lore.

Infrastructure testing is the ‘Green Beret’ discipline in IT, generalists who are equipped to understand whether a new project will be fit for purpose and operate as intended. They can bring a unique focus to reviewing the potential risks of a project, and increase the likelihood of a quality outcome in the increasingly complex and risk-averse world of IT.

www.qualit.co.nz