Cloud First Has To Be A Good Thing
Seven ways to make the most of Cloud First
It had to happen, of course. As reported here last week (LINK: http://www.cloudpro.co.uk/cloud-essentials/public-cloud/5561/g-cloud-iii-goes-live-government-confirms-cloud-first-strategy), the Cabinet Office has (finally?) mandated that cloud providers should be given first bite of the cherry.
We’ve also questioned (LINK: http://www.cloudpro.co.uk/cloud-essentials/public-cloud/5564/cloud-first-policy-will-it-really-make-difference) whether the policy can be quite as effective in the short term, as our illustrious leaders would like it to be.
Will it work? We hope so, we really do. The potential benefits of cost reduction, improved efficiency and so on have the potential to affect taxpayers directly in the wallets, as well as helping improve public services. With this in mind, we offer our top tips on how to ensure the success of Cloud First.
Deal with the physical. “Cloud” is such a nebulous term – well, it is – that it starts to imply that the physical elements of information technology no longer matter. This couldn’t be further from the truth – network bandwidth and latency, storage allocation and proximity to processing, suitability of workloads, all benefit form an understanding of what’s going on under the bonnet.
If we could suggest where to start, it would be to review network sizing and contracts first, based on a forward, realistic view of data transfer needs across all expected services. A lack of network headroom and a restrictive contract could bring the whole thing juddering to a halt.
Remember the hybrid. It might be Cloud First but existing systems are not simply going to vanish in a puff of smoke, for a number of reasons. First, it simply won’t be possible for many legacy applications, which were never designed for cloud-based models. The costs of migration in many cases will be greater than the benefits, and indeed, in some cases there will simply be little reason to disrupt an existing system which is running just fine, thank you.
Most, if not all cloud suppliers are advocating, if not acknowledging the fact that some technology capabilities will remain in house for the foreseeable future, resulting in hybrid architectures that incorporate both hosted and in-house systems. From the perspective of the public body, this also means that the IT department is not going away any time soon.
Upgrade skill sets. Building on this point, cloud will not suddenly render in-house IT skills redundant – far from it. However there is another reason why public bodies need to upgrade skill levels – to take into account everything that the cloud makes possible.
The massive amounts of processing and storage now available ‘in the cloud’ create opportunities to develop new applications and services, for example based on massively scalable, ‘NoSQL’ databases, that can have a significant impact on public service delivery. Authorities that understand what cloud enables will be in a better position to spot opportunities to innovate.
Sort out cost attribution. Shared service models of any form – not just cloud – will always make sense in principle, due to the overheads of procurement. Why buy the same thing three times when it can be bought once?
The challenge with pay-per-use type models associated with cloud comes with working out who actually used a service, and then how to ensure the right costs are attributed. While technology can help, the real challenge is having the right roles, responsibilities and processes in place.
Consider the data risks. The way the cloud handles data is creating a minefield, with challenges including aggregation and its potential impact on personal privacy as well as simply keeping practices aligned with legislation and governance requirements.
We use terms like ‘data sovereignty’ to suggest that as long as we know where data is, we are better able to meet requirements. However this is only part of the story. All public bodies have a responsibility to assess the risks and ensure that the way in which they are handling data is in the public best interest, whether or not current legislation is sufficient (LINK: http://www.cloudpro.co.uk/saas/5330/cloud-society-facebook-analytics-just-tip-iceberg).
Take responsibility for due diligence. Cases like the collapse of hosting business 2e2 illustrate that a listing in the G-Cloud catalogue doesn’t give a supplier some kind of magic wand to protect against business failure. Given the impetus to work with smaller (and potentially less resilient) suppliers, this could be a growing issue.
Of course one solution is to avoid the smaller suppliers but this will not help either government policy nor the SMB sector. In fact the best option is to retain a level of due diligence in the procurement process, incorporating not just factors such as financial stability but also taking into account the fact that many smaller service providers are running on, and therefore backed by the service levels of larger incumbents.
Above all, be the boss. Of course, public bodies and local authorities can simply wait for suppliers to come up with the right answers. The real opportunity cloud presents – as often dressed up in terms such as “time to deployment” – is that new services can be tried, developed, upgraded and replaced with relative ease.
It’s not hard to see the potential chaos that could ensue if an organisation – public or otherwise – is allowed to adopt services at random, without any over-arching strategy. Equally, the benefit of the cloud model to organisations that are looking to become more agile, to innovate, to benefit from new services and even offer services to others, is immense.
The bottom line is that the success of Cloud First is inextricably linked to the public sector’s ability to make the most of it. While this may sound obvious in principle, in practice it requires public organisations of all sizes to internalise that there is more to cloud adoption than picking services from a catalogue. Get this right, and we all stand to benefit.