|
The world of client computing within the enterprise is in the midst of reevaluation. Not only are user needs increasing in diversity and complexity, but the increased capabilities and mobility being made available to them are creating security risks and causing growth in management and support costs for organizations of all sizes. In response to these growing challenges, CIOs and other IT managers are revisiting their client computing strategies and are increasingly considering a move toward a more centralized client environment. To accomplish this goal, a growing number of organizations are beginning to deploy (or pilot) desktop virtualization solutions such as VMware’s Virtual Desktop Infrastructure (VDI) and Assured Computing Environment (ACE). These technologies leverage virtualization software to separate desktop environments from the underlying hardware on which they run, and in many cases, they are enabling greater levels of security and manageability than was possible within distributed desktop environments. All the while these technologies still maintain, to a large extent, the flexibility of the distributed model.
This IDC white paper considers the concept of desktop virtualization, specifically as made available by VMware’s VDI and ACE. The document also considers the benefits and limitations of these technologies as well as the best use cases for each, exemplified through user case studies.
In the early 1980s, the IT landscape largely consisted of centralized computing environments. In this centralized model, the processing and storage of data were performed centrally while user inputs were made from fixed computer terminals at end-user locations. These terminal devices were connected to mammoth-sized mainframe computers located in a remote datacenter that processed the terminal inputs from the user and sent back the outputs to be displayed by those same enduser devices.
As computer software became increasingly complex in both capabilities and hardware requirements, IT departments responded by pushing processing resources further out, moving processing power from datacenters toward department-level minicomputers and eventually out to PCs that sat on individual workers’ desks. This move put unprecedented computing power, flexibility, and customization into the hands of end users, improving productivity and creativity. The invention of the laptop made possible the mobility of those computing resources.
Ironically, this transition toward a decentralized model became universal around the same time that sensitive data was increasingly being stored within the IT environment and users were gaining access to outside networks, introducing concerns around security.
Today, distributed computing environments have become the norm for virtually all enterprise organizations. Although the benefits of this model are clearly understood, many of the costs associated with the model are increasing substantially as environments scale. This is creating pain points for IT managers, including the following:
- Total cost of ownership (TCO). The TCO for an individual distributed desktop ranges significantly but in all cases is fairly expensive. These costs include hardware, maintenance, help desk support, change management issues including application provisioning and patching, and unplanned downtime limiting user productivity. In environments in which PCs are tightly managed, these costs can range from $700 to 1,000 per PC per year, and in very loosely managed environments, they can balloon to several thousand dollars per PC per year. With an approximate 496 million PCs distributed across IT environments, these costs are very substantial.
- Security. The stories about computers either walking off or being lost and exposing valuable data are becoming routine. The potential for this kind of physical theft or misplacement is very high with today’s distributed desktops and, especially, notebook PCs.
- Data loss. With decentralized client environments, information is typically stored and processed locally. As such, a significant amount of data needs to reside on an individual user’s computer, threatening data in the event of a local hardware failure. Even the most rigorous backup policies cannot eliminate this problem. - Regulatory compliance. Governments have enacted laws and regulations
designed to protect sensitive data in order to defend consumers, businesses, and government agencies. Many of these regulations are a result of the drawbacks of a decentralized computing environment and are easier to comply with by moving to a more centralized model. For example, in the U.S. healthcare industry, the Health Insurance Portability and Accountability Act (HIPAA) laws require that no patient medical information be stored on local PC hard drives. With centralized computing models, all data must be stored on the server, making compliance an essential part of the architecture. Similarly, in some regions, such as Japan, regulatory compliance is a primary driver for virtualizing certain components of the desktop environment.
|
