False promises in the cloud

The topic of cloud computing continues to fill the content of IT magazines. The accompanying adverts in the same publications extol its virtues. Any self-respecting IT conference organiser would feel naked if it wasn’t on the agenda. Hardly a week goes by when I’m not invited to a Webinar on the subject. The IT research companies seem to have a production line that is churning out reports and analysis on the take up of cloud computing. Yet the reality out there in user land is that most IT executives I talk to feel the vendors are not walking the talk when it comes to delivering the promise of cloud computing.

Let me say at the outset that I can recognise its appeal. Cloud computing appears to promise the nirvana of utility computing. This is the ability to resource up or down depending on the current business workloads for IT. I know this is something dear to the hearts of most IT executives. They wrestle daily with the challenge of determining just how much resources are needed by business applications. Typically, there is a lot of guesswork in doing this.

Business demands for computer processing power can vary significantly. For example, there can be times in the business cycle when there is a spike in the workload placed on business systems, (e.g. end of quarter reporting cycles). Furthermore, when new applications are installed the IT department has no way of really knowing in advance what impact this will have on the IT environment. In the past the standard response has been to overconfigure the systems to ensure there is sufficient capacity available if or when it is needed. Yet this can leave IT resources sitting under-utilised for long periods and it opens the CIO up to charges of being profligate with the company’s monies. With the costs of IT subject to ever-increasing scrutiny most IT executives are keen to avoid such accusations.

Cloud computing seems an ideal solution to these needs. It conveys the impression that a CIO can pull in extra computing resources as and when they are needed. The capacity management conundrum would appear no more. As with other utilities we consume all you need to do is turn on a tap or flick a switch and that extra processing power is at your disposal.

This is particularly appealing in business right now. The past two years have been a roller-coaster ride for the stock-market. No one can say for certaint that the global financial crisis is behind us and quite a few economic pundits are predicting that the world will experience a ‘double-dip’ recession. In such uncertain times what business needs more than ever is the agility to respond and adapt to whatever are the economic circumstances that eventuate. Cloud computing would seem perfect for these times.

Yet in my dialogue with IT executives I hear a pretty consistent complaint about it. The cloud computing packages being promoted offer none of the flexibility and agility that CIOs desire. Instead, vendors seem to be viewing cloud computing as a variant of the traditional outsourcing service. Their packages are for three or five year fixed periods for the delivery of a specified amount of IT functionality.

The inadequacy of the vendors approach was highlighted for me recently by a CIO in a major distribution company. This organisation has exclusive distribution agreements for leading brand name consumer products from around the world. In such an operating model there is only a limited surety of supply. It is quite possible for the manufacturer of a key product line to terminate a distribution agreement with only a few months notice. Business executives are looking for IT to help them respond nimbly to these challenges.

As such, this CIO began to investigate potential cloud computing solutions. He was particularly keen not to be locked in to any long-term arrangements as he had no certainty what his organisations operating model would look like three years out from now. Instead he wanted some IT service delivery arrangements that revolved around a short term contract that could be adjusted if the business needs suddenly change.

You would have thought his requirements would have been tailor made for cloud computing. Alas, he was to discover this was not to be the case. Instead he found that cloud computing was yet another example of IT suppliers over-promising. His main need was for an IT network delivered within a cloud structure that would allow him to scale it up and down if his organisation opened or closed offices or if it won or lost key distribution agreements.

The response from the telecommunications suppliers he spoke to was that they would provide his network requirements on a fixed service agreement over a three year contract with penalties if he sought to break the negotiated terms. This was exactly what he did not want. The CIO was mystified why a consumer can negotiate a twelve month contract for a service such as wireless broadband but the minimum option for a business is thirty-six months. The CIO in question could have lived with a twelve month contract as that would have corresponded to his budget cycles. However, he found absolutely no willingness to propose twelve month arrangements from any of the telco’s he encountered. As such, he has decided that he has to go back to the old provisioning model of buying the equipment, overconfiguring to address subsequent fluctuations in business needs and running it all in-house.

My IBRS colleague Kevin McIsaac has a good analogy for where he believes cloud computing fits in. He sees parallel’s with our use of cars. Most of us have a need for a family car to handle the majority of our own transportation needs. Kevin equates this to the need for businesses to run their own core business systems. However, Kevin says there are times when we might need to rent a car as we might need a car in a different location or we might need a specialist vehicle like a Ute for a one off job. In addition, we might have a need for a taxi for spasmodic local travel needs when it is not appropriate to drive ourselves. In such cases Kevin equates these uses to when a business might need an external service or cloud computing provider.

I can recognise the validity in Kevin’s analogy. However, taxi and hire car companies succeed because their pricing structure is palatable for their clientele. My observation is that cloud computing companies also need to make their service offerings palatable for their customers. They have to price it so users can leverage the flexibility, adaptability and agility that cloud computing can offer. If they cannot do this then I suspect the IT airwaves will move on from cloud computing in the not-too-distant future and the functionality will go down in the IT record books as something that looked appealing but which failed to live up to its promise.