The launch of an expanded G-Cloud supplier framework earlier this year also marked the formal introduction of a ‘Cloud First’ policy, intended to drive wider adoption of cloud computing in the public sector, which would boost business and offer further savings and efficiencies.
When procuring new or existing services, public sector organisations have been asked to consider and fully evaluate potential cloud solutions first – before considering any other option. Mandated by central government and strongly recommended to the wider public sector, departments have also been told that they remain free to choose an alternative to the cloud if they can demonstrate it offers better value for money.
Government spending on IT remains high and sales from G-Cloud have been rising steadily but not to any extent as to making a significant dent on overall spend. The widespread purchasing of IT services as a commodity was never going to happen overnight.
While the transition to the Cloud is revolutionary the pace we have been expected to move has been of more concern with little time to rationalise why these decisions are being made. Have IT departments had enough time to consider all the potential consequences of letting go of IT-control, governance and decision-making over their ICT infrastructure i.e. servers, networking and storage?
Cloud means the decommissioning of internal infrastructure which on the surface looks like a good plan. Councils save money, reduce capex and are freed to deploy IT support staff elsewhere.
IT departments have now become responsible for just the types of applications and the data and not the management of any infrastructure, which is left to the cloud provider.
The fear of vendor lock-in is also something that is holding us back. Firstly, there is the question of why we need to migrate to the Cloud. As stated, for most it is to reduce the amount of physical infrastructure we need keep and maintain.
Secondly, we need to ask how we are migrating to the cloud. The complexities of cloud service migration mean that many stay with a provider that doesn’t meet their needs, just to avoid the cumbersome and risky process of moving.
As an example, in order to move data from one provider’s cloud environment to another it may be necessary to first move the data back to the customer’s site and then move it to the new provider’s environment.
The problem with this is that the customer probably doesn’t have the infrastructure anymore because they didn’t need it after migrating to the cloud. Furthermore, the data may have been altered for compatibility with the original cloud provider’s system so that what is returned to the customer it needs to be returned to its former state before moving it on again.
In the private sector, we have already seen what happens when customers consider a change; vendors make the whole task of migration more laboured. To put it bluntly, vendors will try to lock-in customers to avoid migration to a new provider. For the customers, who often don’t know how lock-in might impact them it can be devastating.
You may ask isn’t this the same as dealing with any other managed services provider (MSP) or outsourcing partner (SI) relationship – just the underlying IT infrastructure has become invisible. With careful planning it can all be managed and you could move to another cloud or MSP provider. However, there can never be any rolling back to your own internal self managed and controlled IT infrastructure.
Before make a commitment many more questions need to be asked.
What happens if your data is compromised? What if the cloud business model doesn’t stack up as expected? Maybe the experience has been soured by add-on charges and an initial under-scoping of requirements by over enthusiastic sales people keen to secure contracts that allow for future up-selling. What if the provider can’t provide the level of service you expected?
Look at some of recent incidents with large cloud providers and the impact on their clients. In these cases if an organisation decided that it wanted to go back you would be staring a full new infrastructure investment – a task that could take several months, maybe years and significant cost.
There needs to be a sane and critical voice to challenge this “go to the cloud, that’s the future” policy and more understanding of the impact if something went wrong.
The Cloud is revolutionary. It’s a great vehicle to rapidly deploy dynamic infrastructures for specific purposes that align to fluctuating client demands as well as DR and backup purposes.
The advice should be “think it through, plan for all possible scenarios then make a decision based on what fits your organisation best.”
Jacob Barreth is director of data centre provider MigSolv