Skip to content


Thoughts to start a year by

Everyone and their dog seems to be in on the game of making predictions for what is going to happen in the world of cloud during 2013. So after a few minutes cogitation I decided to join the fray. Here, therefore, are six predictions for the cloud sector that I expect to have a growing importance during 2013.

The Tech-Upgrade Cliff

I wrote about this in Business Cloud 9 recently, and will re-publish that piece here shortly,  but I still feel it will be one of the major drivers of a cloud-wards move, particularly for large enterprises, over the next year. At a time of austerity this has one key pressure point – serious reductions in capital outflow.

The Cliff itself will be the cost for many businesses, and large enterprises in particular, to upgrade their current on-premise applications as they either fail to provide the new capabilities the businesses require, or meet their official `end of life’ as a corporate decision made by the softare vendors.

The background to this is the current economic austerity, where many businesses have opted to `sweat’ existing assets to the full rather than invest in applications upgrades deemed marginal or unnecessary. Many of these application are now, however, coming available with upgraded capabilities that the businesses will want to use. Many businesses will, however, still be using versions that are now two or even three steps away from the new version.

They will not, therefore, qualify for the usual discounts on licences they would have been offered if upgrading sequentially and on time. Neither will they be able to upgrade directly from `version N’ to `version N+2’. Instead, they will be obliged to buy – and install – all intermediate versions, and pay full list prices.

Is expected that many vendors will also `sunset’ those older versions of their applications, cutting away all but the most critical levels of support (ie major security patches etc).

Users will therefore face significant direct costs in upgrading, plus all the disruption and risk involved in porting to new versions, implementing required staff training processes and a panoply of other issues.

The alternative will be to seriously consider biting the bullet and moving many of those applications to the cloud. Some of the major software vendors are already offering more of their applications suites as cloud services, which is a good start point, but there are also now a growing number of SaaS services appearing – and continuing to appear – that can match the functionality of the legacy on-premise applications, and provide all the appropriate levels of security required.

I believe the latter half of 2013 will see this trend start to take off.

Start of a `Post Computer’ World

Cloud-delivered services carry with them an important difference – the focus of attention on the `service’ being provided, what is actually required from it and what benefits it brings to the business as a whole. In many ways, the means of delivery is less relevant.

With IT, however, there still remains a great deal of emphasis on the `T’ – technology. This means technology for its own sake.

But this is the year that is likely to change, a change that will bring about the demise of `the computer’ as a definably important `entity’ in its own right. The technology will `disappear’ in terms of its prominence, becoming like a pencil – important that it is available for use by a writer or artist, but of little intrinsic value in itself, if only because of its commodity status and commonality of a design that is fit for purpose.

Computer hardware technology is approaching this point, and software is following along quickly afterwards. There will still be people needed to cut original code – though automated tools to do increasing amounts of this work have been around since Microsoft’s Visual Basic appeared. But the levels of abstraction in what constitutes an `application’ will continue to move away from code per se. The analogy is that we think `motor car’ and think not once about piston rings unless they are not there. That model will increasingly apply to the world of the computer.

A key driver of this is that, as applications and services become more commoditised, they will need to be more widely used – and used by people who know damn-all about `computing’. And that is the way it now should be.

The cloud will play an important part – indeed already does – in the delivery of these services, and its role as the backbone of that process will become the accepted level of understanding that most users will require. As a result, 2013 will see the service provider becoming the base-level brands for most users.

It’s all just `Cloud’

This could, alternatively, be referred to as `Private Cloud’s Last Stand’. One issue that does, I feel confuse potential cloud users more than anything are the arbitrary, and largely artificial, divisions of `cloud’ into sub-groupings. 2013 will probably start with even more sub-groups appearing all aimed at identifying ever-smaller market niches. One of the driving forces of this trend is the need for legacy applications vendors to maintain a market place where they can be recognised and claim `market leadership’.

But as the year goes on I expect to see business users become less concerned about the sub-group of cloud services they belong to and more concerned with the benefits and value they are getting from their services, allowing those criteria to be the arbiters.

Private Clouds will, I suspect, be found to actually add little of benefit to large enterprise operations, and to cost at least the same – probably more – than existing legacy on-premise operations.

This trend will even come to cover cloud-specific divisions such as SaaS, PaaS, social media and the rest. The question will change from `which tech approach is best?’ to `where do I source the most appropriate service for this problem?’

Governance and Data Sovereignty Issues Diminish

This will happen as the ability to create domains of securely managed and controlled environments in any 3rd party datacentre increases, and such tools are now available.

Data sovereignty issues will become a problem of the past. The tools already exist that allow a logical implementation of secure environment A, running on datacentre B in country C, to be re-created, managed and run as secure environment X on datacentre Y located in country Z.

It is certain that the bodies most likely to have trouble with this notion will be the sovereign nation states themselves. Countries like Germany are particularly strict about this. 2013 will see, however, that such strictly applied rules, while sensible in the earliest days of internet-based transaction processing, are now increasingly based on emotion rather than logic.

With German businesses already starting to find it possible to create logical `German’ environments and insert them into datacentres anywhere in the world, real pressure for the reduction in restrictive legislation is likely to follow. The obvious business advantages in terms of cost and operational efficiency, plus greater business flexibility and agility, will add to this pressure.

The IT Department is dead – long live the Service Aggregation Department

The changing job roles and functions of the IT Department have been speculated upon for a while now, but 2013 is likely to be the year when real changes to the roles and organisation start to be made.

This will happen if only because many enterprises will start dabbling with private cloud structures, which will oblige IT departments to start working at provisioning services rather than run and maintain applications. It may not take too long before opportunities emerge to provision services from other resources than internal applications development teams.

Then they will start moving to become aggregators of services, finding appropriate applications and resources, negotiating with those service and resource providers, and using their technical skills to ensure the integration into the services required.

Policy, Monitoring, Automation and Autonomics

2013 will be the year when policy-based automated management services becomes a major component part of all forms of cloud delivery, for without it being extensively applied across the board cloud services will never fulfill more than a fraction of its potential.

To automate service management will, of course, mean continued development of many different types of monitoring tools. Every aspect of cloud service activity will need to be monitored, with the results matched against user-defined operational and performance policies. This will apply across the board, from monitoring and managing operations and access from a security point of view (for example the classic `is that individual allowed to do that, with that data, at that time, from that location?’) through to billing (`that process has been run this many times using that much from the resource pool’).

The policies will also manage the automated responses to monitor inputs to ensure prompt control over all processes in real time. So, in the security example above, if the answer to any of those questions is `no’, the process is instantly stopped and the data defended.

We are also likely to growth in the notion of component self-management as the monitoring and automation technologies develop.

Lastly, there is likely to be growth in the use of autonomics, the software technologies that allow systems and services to learn how to manage themselves. In this way, a process can respond to previously unknown inputs or operations by comparing them with known inputs or operations in order to find the closest match. It can then start adjusting parameters – modifying itself until an appropriate result is achieved. It can then even let other systems know the result if that is appropriate.

This piece was originally posted in Business Cloud 9 on 18-12-12

Posted in Business.


0 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.



Some HTML is OK

or, reply to this post via trackback.



css.php