With the emergence of web-based application delivery - AKA Cloud Computing , AKA 'The Cloud' - software is currently undergoing perhaps it's most pronounced, sweeping, important evolutionary step since the advent of client architecture.

Certainly anything that gets the world's biggest software vendors scrambling to shift their strategies almost volte face is a pretty big deal. Mario Zampi, UK & Ireland Country Manager at application delivery specialists F5, gives iQ his take on the current and future app delivery landscape.

iQ: With a characteristic lack of self awareness, it seems the IT industry has managed to quite literally ‘Cloud’ an important issue – i.e. the sweeping, seminal changes currently taking place in software – with the mixed messages and confusing jargon currently surrounding Cloud Computing. How do you see it?

Mario Zampi: Cloud Computing may be the current term of parlance but that’s really just the latest catch all phrase for an ideal that’s actually been around for quite a while. A less complex way of looking at it is simply as IT resource delivered via some form of web connection (from “The Cloud”) rather than IT that emanates from kit resident in house.
Whatever you choose to call it, the important thing it is proliferating and it will proliferate faster and further as more and more companies come to understand and buy into the idea and begin seeing the Cloud’s potential for improving not only their IT, but their business operations.

iQ: So it’s partly a question of semantics then, as much as the actual shift from the delivery of custom applications to web-based applications? 

MZ: That applications and application delivery are becoming more web-centric is not, in itself, that big a shift or that big a surprise. The vision of delivering applications as a utility via the web isn’t a new one. Look at say, McAfee. It started delivering elements of its security products – online updates and patches and so on – via the web many years ago. 
However, while the ideal and the benefits existed, the infrastructure needed to make the whole thing fly on a wider scale – to properly deliver and maintain entire applications via an online resource – simply didn’t. In essence that’s what’s really changed. The entire back-end has matured in recent years so there’s more stability, more flexibility, greater benefit, and a much wider range of options for companies considering the adoption of a more web-centric approach to application delivery. 

 
iQ: What other factors do you think are making web-based delivery more compelling and attractive for potential adopters? 

MZ: The ongoing development and maturation of software has made a big difference too.

Where applications used to be very much stand alone entities that operated in isolation, for instance, they have a much greater capacity to interoperate and intertwine with one another now.

Take the likes of voice and data and unified messaging. Thanks to IP these systems can now talk to each other to a greater degree than was even thought possible a few years ago and that’s really helping businesses to see the potential of IP-based delivery.

iQ: Where is the market right now then? Does Cloud Computing remain an abstract concept for the most part or are businesses really starting to get it? 

MZ: As far as we at F5 are concerned web-based delivery is already something that companies very much understand, and something that they’re very much buying into and demanding (more and more) as they seek new ways to escape the expense and inefficiencies of the traditional application and application delivery model.

As such, I’d say that it is already very much a pull market – especially in the mid market and up into the high-end enterprise space.

iQ: How should businesses largely unfamiliar with the area best think about and approach “The Cloud” then? It’s just such a nebulous, non-specific phrase...

MZ: In a lot of cases it is first a question of trying to put an end to the perpetual firefighting that’s so common in modern organisations and in IT shops in particular. You need to take a step back, take a breath, look at the bigger picture and think about where your business could benefit from an application predicated as a utility rather than as a traditional, finite, owned, datacentre-based tool.

For example, it makes perfect sense that those organisations based across multiple sites in multiple countries with multiple-connections for multiple user types, should be (and are) beginning to look at the bigger picture from an application delivery standpoint.

It’s a question of realising that one size need no longer be made to fit all; that the kind of traditional custom applications that we’re all so accustomed to, and that so many businesses default to, are often needlessly expensive, unwieldy, inefficient, and, just as importantly, no longer necessarily the best answer.
 
iQ: There seems to be a great deal of misinformation out there however – particularly at the intersection between Cloud Computing and Virtualisation – and end users could surely be forgiven for ending up confused?  

MZ: Virtualisation is just another part of the puzzle – albeit an extremely important one
– but it again, it’s important to view it the context of a more expansive holistic picture.

The problem here though, is that many current virtualisation projects and providers remain blinkered and only deal with virtualising one or two parts of the environment – and often in almost complete segregation from everything else.

Piecemeal virtualisation or virtualising in isolation in this way really isn’t a good idea. It’s imperative that businesses think about how the knock on effect of virtualising one area of
IT operations will impact dozens of others. Again it’s about taking that step back.

iQ: Is it again a case of confusion and misinformation prevailing over clarity and crystallised joined up thinking? 

MZ: The situation isn’t helped in the current market landscape where, of those businesses that are familiar with the concept of virtualisation, many seem to equate it with a single vendor – namely VMWare – which, whilst a hugely important player, is simply a key part of that bigger jigsaw where F5 is concerned.  

iQ: So it can come down to whoever shouts the loudest?

MZ: From our point of view at F5, virtualisation shouldn’t be about any one single discipline. It’s not just storage virtualisation, network virtualisation, file virtualisation, server virtualisation – it’s about all these and more. In this sense virtualisation is, or should be, just a means to an end – one tool of many that help manage data growth, spend, and management.

It shouldn’t be viewed as a panacea. And the case for virtualisation has to be every bit as much a business argument as it is one based on technology.  

It is also about recognising the need for change from an application delivery and storage perspective. Look at storage virtualisation. It’s an increasingly common practice and yet the majority of companies continue to miss a major trick in that they’re not virtualising in one vital area – their unstructured data.

Whether you are an individual or a commercial business or – as is the case with virtually every IT user on the planet – both, you will be using certain applications and creating more and more files in those environments every single day. Millions and millions of unstructured files that combine to create an expensive, inefficient data nebula.
Common or garden storage virtualisation isn’t enough here. To properly rationalise storage time and costs, the business has to look at virtualising its unstructured file data too.       

iQ: It’s obviously still early days for the Cloud/Utility/Virtualised computing model, but as the technology matures and moves on could we even see the tail starting to wag the dog? – To see the application delivery method start shaping applications themselves?   

MZ: I’m not sure and it’s way too early to say – although it would be great news for us! – but I do think we’ll start seeing the emergence of much smaller more specialist applications, that these will begin to proliferate in numbers, and, conversely, that the use of more traditional, weighty applications will begin to decline.      
The migration from fat to thin client models will certainly continue moving forward and we’ll likely see datacentres slim down too as these smaller apps gain users, popularity, and serious market traction.
The long and the short of it is that there is a great deal for businesses to think about and consider, and that they have to start weighing it all up now – to take that step back, look at their core business drivers and see how best applications should be being delivered to meet them.  

Things are changing fast and my advice for IT and business management is this. If it’s in app – and especially a business critical app – then gen up now and gen up well.