No Jitter is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Enterprise Cloud Visions

"This isn’t about evolving to the cloud. It's about evolving our IT model. The cloud is just a kind of resource model."

It seems like every time we think we understand something about cloud computing, somebody moves the bar. An example is the "as-a-Service" categories; SaaS, PaaS, and IaaS. Those are the kinds of cloud computing, right? Well, according to my latest talk with enterprises, it's not right. They're saying that there are two meaningful visions or models for the cloud, and "aaS" isn’t part of either.

Most enterprises are a bit frustrated by the cloud computing debate because it often misses the point from their perspective. "This isn't about evolving to the cloud," one told me. "It’s about evolving our IT model. The cloud is just a kind of resource model". That seems to cut to the heart of the enterprises' issue set. They have IT commitments and they'd like to integrate these with cloud computing to the extent the integration makes business sense. But they have to start with what their IT commitment is, and that's the topic they feel is being ignored. Enterprises seem to divide themselves into two classes of IT commitment; virtualization-centric and service-centric.

Virtualization-centric enterprises often developed their IT resources by deploying application-specific servers and software as needed, and this created a rather decoupled IT model. Often they did some M&A that resulted in the integration of multiple new servers and applications. This kind of user is likely to employ GUI tools to assemble screens from multiple applications to create a composite view. They're likely to be looking now at server and data center consolidation to create a more efficient IT infrastructure, and they're very interested in virtualization.

According to enterprises I talk with, their "cloud strategy" is to create a hybrid cloud that combines their own data center virtualization initiatives with cloud hosts for virtual machines, giving them a combination of a data center backup strategy and a way to use public cloud resources as an overflow server pool during periods of heavy use. Because they've already committed to virtualization, they're most likely to be looking at something that connects their environment to the cloud (so, for example, they're very interested in VMware's vCloud Connector).

Service-centric enterprises are most often companies who have embraced a single software vendor for most of their platform needs, and most of them have also accepted the service-oriented architecture (SOA) model of componentized applications. Many of these users have a small number of large computer systems in their data center, and are using multi-tasking and directory tools for load balancing and redundancy. The majority of these users aren't looking at consolidation per se, but they are interested in application reliability and availability during periods of data center failure or congestion.

Some of these players like the virtualization approach, and this group may also be thinking of public cloud services as a source of virtual hosts for their virtual machines. The majority, though, are looking at hosting applications more than virtual machines. They're thus very interested in platform-as-a-service offerings like Microsoft's Azure, where they can make their cloud resources look (again, via a "connector") like an extension of their own data center. Some of the service-centric types even see public cloud "services" (as in Software-as-a-Service) being integrated with their own applications via the same connector.

The service-centric group is likely to be more involved with line departments in managing their evolution to the cloud, according to enterprises. In fact, this group has nearly three times the chance of creating cloud project goals and managing projects using a combined IT/business team than the virtualization-centric group. The difference seems to stem from the latter's "consolidation" motives; server consolidation is typically an IT organizational goal rather than a business direction.

Both groups are good prospects for service provider cloud services, but the virtualization-centric group seems to have advanced their consideration of the service provider cloud farther and faster than the other group. Part of this seems to be due to the early provider emphasis on infrastructure-as-a-service, which is easily adapted to the notion of creating connections between cloud-hosted virtual machines and virtualization in the data center. Another factor seems to be the success that the IT giants like Microsoft have had in the service-centric community; that group accounts for most of their early sales.

What both types of enterprise would like to see is a strategy that converges public cloud technology and "cloud connectors" to serve either mission. Nearly all the enterprises either have some applications that fit both models already, or think they will develop them. They're concerned about differences in how public cloud computing services support the two models creating management and integration issues for them down the line. IBM is seen by enterprises as the most likely to create such an architecture, with Microsoft a close second and Oracle third. Among users with strong UC/UCC focus to their early projects, Microsoft is the clear leader.

Almost all the enterprises I surveyed (277 in total) said that they believed that most cloud service vendors did not understand the enterprise's IT goals, didn't understand these two models of IT/cloud planning, and offered no real help in creating a cloud business case. "Don’t tell me that 'the cloud is the way of the future' and expect me to sell that to my boss," one complained. So, hopeful cloud vendors, maybe you should review your messaging!