Getting virtualization right: Five virtualization rules every organization must know

Saroj Agrawal and Vaibhav Mishra

MBA-IT

Indian Institute of Information Technology - Allahabad

Technology organizations have one mission i.e. to provide technology-enabled services to their businesses which helps in lowering cost and risk and provide higher service levels. Virtualization has a key role to play in this process whose outcomes are dependent on several key factors. From decades of experience working alongside businesses of every shape and size and in every economic climate imaginable, the five common virtualization rules has been identified, that are employed by organizations. The technology differs in complexity and scale from one business to the next, therefore nearly every successful organization deployed the same five rules to realize the full potential of virtualization. The rules are as follows:

1. Focus on connections

2. Manage one infrastructure

3. Automate to enable scalability

4. Think beyond the data centre

5. Be flexible

Focus on connections

The first rule is to focus on the connections between the applications, servers, storage, and network. In general, organizations initially take a technology - centric approach to virtualization. The “connections” rule requires some rethinking about how a virtualized server, with its greater mobility and speed, interacts with the rest of the hard‑wired infrastructure. Just because a virtual machine can be provisioned or moved quickly does not mean that the connections to storage and network will automatically follow. Broader view of virtualization includes all the elements of a service and most organizations ultimately find its importance because any gains on the server side will be muted by overhead costs that spill over to the storage or networking side. With a focus on connections, mobility and speed of the entire service infrastructure is designed with virtualization in mind. There are key technologies that enable you to transform your hard-wired infrastructure into a business‑ready state. Virtual connect technology virtualizes the physical connections between servers and storage as well as servers and networks. By virtualizing these connections, Virtual Connect allows server managers to wire everything once, and then add, replace, or recover servers without disrupting the network or storage topologies. The result is that new deployments or moves can be accomplished in minutes, rather than days. Another advantage of Virtual Connect comes from the Flex-10 module, which flexibly allocates 10 GB Ethernet bandwidth across four NIC connections. This removes the need for additional costly NICs, switches, and cables, while increasing bandwidth flexibility. As a result, we can save up to 75 percent in network equipment costs.

Manage one Infrastructure

The second rule acknowledges that with new technologies come new tools. It is all too common to find virtual infrastructures being managed with one set of tools and physical infrastructures being managed with another. This dual-infrastructure approach leads to multiple sets of operations tools, management data, and processes. If routine operational requirements, such as service availability, event management, problem isolation, and compliance reporting, were difficult before virtualization, imagine the cost and complexity increases that are possible in a virtual physical divide. Technology-enabled services will always be made up of physical and virtual components—applications, middleware, servers, storage, and networks. This makes monitoring and managing this mixed environment in a unified manner essential for technology providers to fulfil their mandate to the business. The key is holistic management of one infrastructure. The solution is Universal Configuration Management Database (UCMDB) that provides a consolidated repository of the physical and virtual infrastructures and services. The Discovery and Dependency Mapping process not only automates the population of the UCMDB, but it also maps the relationship of these logical elements to the overall business services they support. Using the capabilities of UCMDB, one can proactively model the service impact of proposed changes and track configuration changes to all infrastructures, including virtual elements.

Automate to enable scalability

The third rule is about growing a virtualization footprint, reliably and cost-effectively. Automation allows technology organizations to scale services up or down quickly, without requiring corresponding investments in staffing costs. It also takes the element of human error out of complex manual tasks. For these reasons, wide-scale adoption and success of virtualization is dependent on automation. Automating routine, operational, and end-to-end processes is essential for driving out service costs, increasing quality, and maintaining compliance. Automation can be deployed at several levels. At an infrastructure level, the Data Center Automation Center can automate and orchestrate provisioning, updates, and compliance tasks across an entire hardware stack—including network, server, storage, and even client computing devices. Automation can also be applied to daily maintenance tasks, such as backup and recovery. The Data Protector Software automates high‑performance backup and recovery from disk or tape to provide 24x7 business continuity, while lowering the cost of doing so by up to 70 percent. Automation of data collection, correlation, and analysis for availability and performance data can also be enabled - as virtual machines are put up, taken down, and moved from host to host. Without automated data collection, it becomes difficult to make simple decisions on what workloads are compatible and how many virtual machines can fit on a host. This can lead to incorrect configurations and a notably negative service impact. Having good decision support data makes automatic provisioning of servers more efficient and less error-prone, which leads to improved service levels.

Think beyond the data centre

The fourth rule focuses on security, reliability, and leveraged investments through client virtualization. Virtualization does not end at the data centre. Many of the investments made in the data centre can be leveraged into virtualized desktops to reduce the risk of unsecured endpoints and data as well as improve employee productivity, while lowering the overall cost of services. Extending virtualization to the desktop makes the business more secure, agile, and adaptable to your workforce at a lower total cost of ownership (TCO). Users become more productive through vastly increased flexibility with a more reliable local desktop experience from anywhere they may need to work. In addition, administrators can remotely support, troubleshoot, and manage hardware, operating systems, applications, and services for a broad number of users, without the need to crawl under desks or visit distributed desktop locations. Client virtualization solutions can keep data in the data centre securely. Using remote client virtualization solutions, the visual desktop experience is delivered from the data centre to the user’s thin client or access device. This can reduce the risk of data theft from end-user devices significantly and protect against unauthorized access, making it easier for your business to meet compliance requirements that address proprietary data or customer information. Implementing client virtualization can also make it easier for you to meet both your organization’s and your customers’ environmental policies. The small footprint of thin clients requires significantly less component and packaging material usage. And the overall lower weight translates into less fuel consumption during delivery than a standard PC. Also, thin clients have no moving parts—no fans or hard disks—making them quieter, more reliable, and more energy-efficient than conventional PCs. Reducing ambient office noise can reduce fatigue and stress, making workers more productive. In addition, thin clients generate much less heat, so offices stay cooler with lower air-conditioning energy consumption, resulting in lower electricity bills and an overall reduction in CO2 emissions. The use of thin client technology, in parallel with the energy-efficiency measures in the data centre, means that the carbon impact of the end‑to‑end infrastructure can be reduced—with overall improved efficiency due to a reduced service load. A thin client and Virtual Desktop Infrastructure (VDI) session generally uses below 30W of electricity combined—even under heavy load. A desktop may consume more than 150W under the same circumstances as a VDI session. Just add up the costs using 100, 1,000, or 5,000 desktops at the current electricity rates, and think about the potential cost savings. This kind of savings can affect a company’s bottom line. Experience shows that, in order to deliver the security, compliance, and cost benefits the business expects, the client environment must be flexible enough to match the computing needs of the employee base. These needs vary greatly by position, by department, and even by location, project, or time of the year. Client Virtualization solutions help to enhance the reliability and security of the end-user environment, while simplifying management. For most organizations, when it comes to virtual client technologies, one size does not fit all. Targeting the right solution for the right user group or environment allows the best performance at the best cost. For this reason, many enterprise businesses have deployed, and will continue to deploy, multiple client virtualization technologies for their user environments.

Be flexible

The last rule concerns how you source the skilled resources needed to put the other four rules into play. In the face of budgetary constraints, successful organizations have shown that getting creative with sourcing models is one way to fill in the skill gaps that often prevent broader adoption of virtualization. Each organization’s ability to realize the full potential of virtualization is based on its specific needs and resources and its familiarity with all facets of virtualization. Many service based companies provide flexible service options designed to help each organization develop and execute a service-centric virtualization strategy, tailored to its needs. The services provided can help to assess, design, transition, operate, and improve the infrastructures, applications, operations, and client architectures continually. Alternatively, they also provides virtualization outsourcing services that can help to plan, build, and run an outsourced virtual environment, designed to meet your organization’s goals.