Ovum’s Virtualization and Cloud Management whitepaper starts with an assertion that InsightaaS has often made as well: that virtualization “has expanded well beyond its original value proposition” — described by Ovum as being “simply increasing the level of server utilization,” and by InsightaaS and others as being tied primarily to the consequent ability of delivering additional compute resources in CAPEX-constrained environments — to a point where virtualization…now provides solutions for desktops, applications, servers, storage, and networks.” This expanded scope gives rise to a wide range of management challenges, which are further exacerbated by cloud “adding another dimension” of complexity to the task of delivering seamless service to end customers.
Ovum believes that this path may lead to an unwanted destination. Unless “all the elements [present within the many virtualized and cloud technology stacks] can be unified and managed from a single control point,” “virtualization and cloud computing…will increasingly become vulnerable to fragmentation within the enterprise.” This challenge is especially acute in cloud: Ovum positions orchestration as “the new fulcrum for management and control…at the center of a federated and fragmented IT supply chain, where applications, services and infrastructure are separated by location and ownership. While the nirvana of the cloud world is for any workload to execute in any location on any platform, the reality is that the different standards make this something that is currently only possible between technologies that have established working relationships.” The reality, Ovum states, is that “the move to the cloud for many organizations is a journey that involves managing a heterogeneous environment in a state of flux.”
What are the keys to successfully navigating these shifting tides? In a thoughtful section of the paper entitled “The management layer is now assuming the responsibility for providing interoperability between technologies,” Ovum highlights three key issues:
- Server performance and availability. Ovum offers an intriguing view of the points at which hypervisor technologies lock customers in, and defeat the objective of managing heterogeneous environments from a single console, highlighting management tools and APIs as chokepoints requiring attention. A second key performance and availability issue is virtual machine sprawl, which consumes resources, complicates software license compliance, and creates security risks.
- Application virtualization. Ovum describes different approaches to software virtualization and the increasing use of virtual appliances (VAs) with different levels of capability and integration — which leads in turn to a need for IT departments to adopt IT Service Management (ITSM) principles to ensure that they use a framework for integrating key capabilities. Use of VAs poses a unique challenge for IT: the VA is “a hybrid solution where an application is installed on top of the independent software vendor’s preferred operating system.” Typically, Ovum finds, the OS used delivers “the bare minimum required to execute the application.” From a user perspective, this represents simplicity, since “this combined application and operating system is then contained in a single wrapper, and the end user only needs a virtual player to execute the application.” The issue for IT stems from the fact that “in some VAs, an embedded hypervisor is also provided.” The embedded hypervisor eliminates the need for a virtual player, but since “currently only one hypervisor can be installed per physical server,” the use of “hypervisor-embedded VAs requires a deeper knowledge of the infrastructure.”
- Hybrid clouds. Ovum notes that “cloud has become segmented…another source of IT supply.” According to Ovum, though, this segmentation has not been reflected in IT architectures — there was a presumption that cloud would “sweep away existing infrastructure and practices,” and as a result, it was not deployed with an eye towards maximizing service delivery cost/value. This situation is changing as vendors introduce chargeback capabilities, but as Ovum says “to fully exploit these techniques, the IT departments need to have successfully implemented, and consistently utilized, IT accounting processes.” Achieving necessary levels of simplicity and perceived fairness in these chargeback systems is challenging — and critical — if the full cost/benefit of hybrid is to be understood.
Like the Forrester paper reviewed elsewhere in this series, the Ovum whitepaper concludes with an evaluation of vendor positions. Like Forrester as well, Ovum positions HP as the most advanced supplier in this space. HP, Ovum believes “understands the business perspective of cloud and virtualization management.” Ovum’s commentary helps illuminate the factors that it considers important to effective supply of virtualization and cloud management technology: it lauds HP for being “particularly strong in financial management…[also having] leading scores in virtualization management, and reporting and integration.” Additionally, the analyst firm states that “HP was the clear leader [amongst 11 firms evaluated] in execution dimensions: interoperability, innovation and scalability.” It’s difficult to say whether enterprise customers and SMBs will be more inclined to find value in management systems and reporting, or in the more ethereal categories of innovation and interoperability — but it is safe to say that firms in both categories will attach value to at least some of these attributes.
Note: this is one of three reports reviewed in the “Demystifying private cloud” series. To access a guide to the material covered in the series, please click here.
To access the Ovum Virtualization and Cloud Management report, please click here