A Guide to Managing Private Clouds
April 19th, 2010 By: Industry Perspectives
Richard Whitehead is a director of Data Center solutions within Novell. He has also held senior positions in product management and marketing for Citrix, Franklin Covey and WordPerfect.RICHARD WHITEHEAD
Cloud computing is an enticing model, promising a new level of flexibility in the form of pay-as-you-go, readily accessible, infinitely scalable IT services. Executives in big companies and small are embracing the concept. But many are also questioning the risks of moving sensitive data and mission-critical applications into the cloud.
There are hybrid models emerging that offer some of the benefits of cloud computing without the pitfalls. These models incorporate what the industry is calling “private clouds.” These private, or internal, clouds house all of a company’s internal data and applications but at the same time give the user more flexibility in how services are rendered.
Technology is here to facilitate this more practical move into the cloud. Following is a guide to how private clouds work and an explanation of how they can be managed using service-level agreements (SLAs) and business service management (BSM) technologies.
What is a private cloud?
The term “private cloud” refers to a proprietary computing architecture that provides hosted services to people behind a firewall. In a 2009 report, Forrester Research analyst James Staten refers to internal clouds as “pools of virtual machines (that) can be built upon either virtual server or high-performance computing (HPC) grid foundations and can be operated according to the specific security and process requirements of the business.”
The move to private clouds is part of the evolution of the data center, from a basic warehouse of information to a smarter, more agile deliverer of services. Data centers started out as self-contained entities and grew into sprawling megaplexes as companies added computer after computer after computer. Virtualization has consolidated the physical server sprawl, helping companies save on real estate, power and cooling costs. But it creates a new challenge – managing all of the physical and virtual servers. It is harder to manage what you cannot physically reach out and touch.
In the virtualized world, IT shops are starting to think about in terms of packaging “workloads,” not just running applications on physical servers. Workloads are portable, self-contained units of work built through the integration of the operating system, middleware and application. They’re not stuck in the data center rack; they’re mobile and can be moved across physical, virtual and heterogeneous environments.
A group of workloads is known as a business service. This approach represents a fundamental shift from managing physical servers and applications to managing business services composed of mobile workloads that can be mixed and matched in the best way to serve the business. Managing IT to business services, a concept also known as the service-driven data center, is becoming a business best practice. IT is at a stage where companies see how the service-driven model can work but are still figuring out the execution – similar to the way it took about a decade before retailers mastered the art of e-commerce.
The biggest advantage of the service-driven data center is that business services can be dynamically managed to SLAs, and moved around as appropriate. This gives companies the ability to attach processes to services by connecting workloads to virtual services. Now, for the first time, a business process is logically connected to the hardware implementing that business process. Even more importantly, in the middle there are mobile workloads that ensure the business service meets SLAs – in other words, the business service can be managed independent of the hardware. Business services can be managed to SLAs because they are not tied to the business server and can be moved around as needed to meet SLAs.
To meet a business’ requirements, IT can call on another set of tools – business service management (BSM) software. BSM is important because it connects business processes. For example, in an e-commerce environment, it would connect order entry and fulfillment. How fast are orders getting through the system? BSM ensures services are connected to the workloads and virtual servers supporting the service. If you have a problem, for example, a virtual or physical server problem, it can be easily moved.
With the service-driven data center, an organization can achieve a true private cloud – one that is virtualized, workload-based, and managed in terms of business services. Workloads are managed in a dynamic manner to meet business service-level agreements. There is a clear, logical progression from physical server to virtualization to the workload to business service to business service management.
Cloud computing is still abstract and intimidating to many enterprises, and probably will be for some time. But companies can begin taking small steps toward full-blown cloud implementations by adopting private clouds. Managing their data center to SLAs using business service management tools can help organizations realize the benefits of private cloud computing. And, over time, enterprises can federate their private clouds with external clouds.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.