Thursday 15 October 2009

Assessing the maturity of cloud computing services

The number one challenge in cloud computing today is determining what it really is, what categories of services exist within the definition and business model and how ready these options are for enterprise consumption.

Forrester defines cloud computing as a standardised IT capability (services, software, or infrastructure) delivered via Internet technologies in a pay-per-use, self-service way.

While definition is crucial to having a fruitful discussion of cloud, the proper taxonomy and maturity of these options is more important when planning your investment strategy.

To this aim, Forrester has just published our latest Tech Radar that maps the existing cloud service categories (not the individual vendors within each category) along maturity and value impact lines to help you build your strategic roadmap.

The report identified 11 service categories that fall into three classes of cloud services – software you rent (Software-as-a-service, or SaaS), middleware services and platforms that help developers build cloud-based applications, and infrastructure services and platforms that are places to deploy cloud applications.

Note that we did not restrict the second and third categories to Platforms-as-a-service (PaaS) and Infrastructure-as-a-service (IaaS) platforms – because there is value in discrete middleware and infrastructure services in the cloud just as there is in your datacentre.

While Amazon Web Services is clearly an IaaS leader providing a robust set of compute, storage and middleware services, there are also discrete offerings, like Boomi’s cloud integration service that does no more than integration and does not need to be a full platform player to be valuable. In fact there can be significant advantage that comes from this type of focus on just one thing.

Oracle’s Larry Ellison continues to grab headlines for his assertion that the cloud isn’t anything new and to his credit he’s at least half right.

Just because applications are delivered from a cloud infrastructure doesn’t mean any of the aspects of application design or components of a service-oriented architecture don’t apply – in fact they have just as much relevance as they did on-premise.

But how these services are delivered is what is different. There is a clear difference between an on-premise, single instance deployment of TIBCO and the highly scalable, multi-tenant integration service from Boomi.

Tenancy, shared economics, virtualised deployment, and cloud service-to-service integration are game changers for those who are using these services. Not only do they bring potential cost advantage to applications that might otherwise have been deployed on-premise but they create opportunities for new business applications that simply wouldn’t be feasible any other way.

That’s clearly the case for Cryoport, a maker of cryogenic containers for live medical tissue.

Delivering these materials safely was very hands-on and expensive until cloud computing came along. Same with genome research; medical research universities and pharmaceutical companies had to make large investments in HPC labs to crunch the massive volumes of DNA data, until cloud computing economics came along.

But cloud computing isn’t the be-all, end-all that many portray and they can’t suit all uses today and may not in the future either. Thus it behooves you to build your roadmap using guideposts to the maturity and applicability of these emerging options.

We hope this report helps you bound and plan your cloud investments and look forward to your feedback on how we can assist further.

Original Article - http://www.computerworlduk.com/community/blogs/index.cfm?entryid=2563&blogid=23

No comments: