What Does Cloud Computing Mean to You? - An Interview from Two Perspectives
By Morgan Killick
In the first of a series of two articles on cloud computing, Morgan Killick of ESP Projects and Stony Grunow of Third Sector IT examine what cloud computing is, the benefits and drawbacks. Their writing styles and focus are different, but together we hope they will give a comprehensive and balanced picture of the issues surrounding cloud computing.
What does cloud computing mean to you?
Cloud computing for me is about the outsourcing of IT functions to third party service providers i.e. companies offering subscription-based services to you via the internet. Although this is by no means the ‘dictionary definition’ of the term, I think it accurately describes how our clients ‘encounter’ the cloud – as an increasingly attractive alternative to in-house systems for document storage, databases, email and collaboration tools and other functions that have traditionally been delivered by a server or servers in their offices.
I like and agree with Morgan’s definition, so I’d like to offer a tangible metaphor to complement the abstract. I view the journey from traditional to cloud computing as something analogous to the transition from owning a bicycle (with its capital and maintenance cost) to using London’s Boris / Barclay’s Bike scheme, where I have on-demand bike use for a service fee. Or going from owning a car to renting one or using Streetcar/Zipcar. Or going from having a household electricity generator to just plugging my lamp into the mains and paying my bill a month later. While the IT world has its own unique characteristics, this transition is something people are instinctively familiar with.
I should also add that there is still a broad range of services offered, from Amazon’s EC2, which merely outsources the server, to web-based Software as a Service (SaaS) which is what I find most interesting and relevant for charities today, and what I generally refer to in this discussion.
I see three principal advantages:
Reduced capital expenditure
Clearly, servers are expensive pieces of equipment, and are typically in need of replacement every 5 or 6 years. In some cases, Cloud services can negate the need for these servers. But it should be remembered that this benefit only rings true when organisations are either thinking about replacing an existing server or when buying a server for the first time – at other times, costs taken in the round may not be that dissimilar.
The second and perhaps more distinct advantage is flexibility – in-house systems are (rightly) ‘specced up’ according to immediate and predictable need. A good IT consultant will talk through what your requirements currently are and how they might change, in order to avoid either under-sell or over-sell. The downside of this approach though is that any big changes to staffing levels or ‘who works where’ may involve additional investment to restructure the system. With cloud computing, there is far greater ability to adapt to such changes - simply changing the number of subscribers to the service without the need for additional resources and expenditure. Moreover, as the services are delivered over the internet, it is much easier to deploy or re-deploy staff in new or different offices.
Reduced IT management overheads
Finally, there is the prospect of reduce costs as the burden of maintaining and fixing servers falls on the cloud provider rather than the end-user. I say ‘prospect’ advisedly – although cutting tech support costs is as attractive proposition for many, in practical terms, most small to medium sized organisations will still need IT support, even if some functions are delivered in the cloud.
Forgive me for dipping into metaphors again!
Business theory suggests that when asking an organisation what they should do in-house and what they should out-source, the answer depends on what their core strengths are.
And if it’s not your core strength or unique selling point, it doesn’t matter if you can do it or not, it’s whether you should do it. Competent IT staff can perform any number of functions, so when offered the choice between patching a server, or deciding on the best way to make an organisation more efficient given future challenges, both are necessary but only one can be done by in house staff.
We could sew uniforms and mill paper, but we don’t because our organisation is leaner, faster, more nimble, and more productive by shedding these tasks and simply picking up uniforms and paper supplies from somewhere else. And to me, that’s where the strength of cloud computing lies.
The main advantage of cloud computing is that it has the potential to provide more reliable, secure, and productive technology tools to all the stakeholders in an organisation while using less resources, employee time, and capital.
No longer do organisations seriously consider buildings their own servers and desktops, or creating their own office productivity suite, because there are more efficient alternatives out there. Reasonable people can disagree if cloud computing is currently at the level of maturity needed to move software from in-house to the Software as a Service model, but whether you think it is ready or not today, it is coming and it will be the dominant model.
Once again, I think there are three categories worth elaborating here:
The internet itself:
Clearly, moving significant functions to the cloud is only an option if you have a fast, stable internet connection. For those without, it would be a crazy choice. We’ve all had that moment of panic where even the best internet connection drops out and we don’t get email and can’t browse websites. In a world where our files and mission-critical databases are also stored online, regular droppages could be nothing short of catastrophic. This isn’t an argument against cloud, its just to re-iterate that it’s not for everyone.
It is often said that cloud services are - de facto – cheaper than their alternatives. In my experience though, you have to take case-by-case approach and engage with each proposal to fully understand the cost implications. Few cost comparisons that I have seen have taken account of the hugely preferential licensing costs for charities for example.
The important thing to recognise, is that all computer systems (unless you are privileged enough to be starting anew) are to a certain extent ‘entrenched’ and changes will therefore carry ‘transition costs’ (such as fairly expensive consultancy work, re-training and the time and money involved in re-engineering current systems) as well as the cost of the cloud service itself.
Even where there is little transition cost, cloud doesn’t always stack up financially. Take online backup for example - there are some fabulous and incredibly cost effective services out there, but the minute your backup set outgrows the entry level price brackets, you are probably going to be paying far more than the cost of an external hard drive not just once, but in every year of usage. As long as that drive is kept safe, it is a far less costly solution over both the short and the long term.
Lack of impartial advice relevant to small to medium sized third sector organisations:
The early adopters and enthusiasts of cloud services have typically been those who stand to gain most from the advantages listed earlier – large corporations with huge server maintenance and replacement commitments, or those highly technically literate, mobile or flexible workers for whom the traditional office-based client-server network was never relevant in the first place.
It is these people that form the vanguard of the push for cloud. But their own experiences are perhaps not so relevant to a whole tier of organisations - with 5-50 largely office-based staff - where the benefits are not so clear cut.
Amongst this client base, with perhaps one or two servers, it is far less likely that adopting a cloud solution will permit the disposal of a server, as that server is likely to be doing an awful lot more than one thing.
Similarly, their remote or home-working needs may well met by existing – and free - services like Remote Desktop or logmein, which can be implemented with less disruption, greater ease and more tailored security than cloud systems.
I see a couple of potential drawbacks:
Change management drawbacks
The first and major drawback is that migrating to cloud computing will involve change, and no matter how great the benefit is, rarely does a change come along that doesn’t cause disruption. Inevitably, some functionality present in the old system is lost, and new functionality takes time to be learned and leveraged. There’s always a valley of pain (image below).
Whether this is worth it depends on your old system, the new system, and and the nature of the organisation / individuals involved. Without knowing these three things (at a minimum), no intelligent recommendation can be given as to the suitability of any new system, be it cloud computing, a low-carb diet, or whether to change what school your children attend.
In an ideal world, the transition from old to new will look something like the function below, with a short drop in functionality upon the introduction of the new system, followed by continued enhanced use.
But the improvement is theoretical until you get there, and improvement may not ever materialise. The business and IT world is littered with stories of failed implementations of CRM or ERP systems (google "erp OR crm failure” for more info) and whether you are implementing a CRM, ERP, or just moving aspects of your business to the cloud, there are risks.
The way to mitigate risks of moving to the Cloud is the same as any other new system: understand, evaluate, get feedback from other users, test, re-test, compare, train, and finally roll-out.
The advantage of cloud computing is that, more so than other systems, it can be evaluated, tested and rolled out incrementally, safely, and at no large upfront cost.
Cloud Computing drawbacks
The oft-repeated theme is that cloud computing requires a constant connection to the internet. Yes, this is true, but it is equally true of the majority of non-cloud solutions. Take any office using traditional, non-cloud computing, cut their internet, and watch panic ensue. Cloud or not, your organisation probably cannot function without the internet. Your modern office will need electricity, lighting, heating, and the internet.
The major drawback of cloud computing is the same drawback as a rental car, or anywhere one pays for a service rather than an asset. You can’t swap out or upgrade the radio or speakers in your rental car, or put on snow tyres. With auto rentals, what you gain in service and zero-capital costs you potentially lose in configuration and adaptability.
The same holds true for cloud computing. If your business process works with the out-of-the-box cloud Software as a Service (SaaS), or is close enough to be modified, you will save time and money and realise massive efficiencies. But you will loose some control, and some things will simply not be possible.
Throw enough money at a non-cloud solutions and you will have all the benefits of the cloud. We aren’t discussing cloud because it can do more, we are discussing it because it can provide high levels of service at formerly unimaginable prices. Ultimately, cloud computing is a subset of the capabilities of traditional computing. If your organisational needs fit within that, consider moving aspects of your business to the cloud, if not, don’t.
- Case Study – Oakleaf flies into the Cloud
- Cloud Computing 101
- Web Based Services - Tales of the Unexpected
- What should third sector organisations be thinking about with regard to cloud?
Published: 29th November 2010
Copyright © 2010 Morgan Killick