Desktop virtualization and application virtualization technologies are increasingly used throughout enterprises to reduce hardware costs, improve manageability, and ease deployment of corporate software assets to PCs.

Desktop virtualization enables deployment of a full desktop environment with applications in it to PCs, and lets them run on a remote server. Using it IT can achieve greater efficiencies by centralizing PC resources.

Application virtualization solutions allow quick and easy deployment of applications to employee PCs and external PCs. They simplify PC lifecycle management tasks and create better efficiencies in using and managing applications on desktops.

Application virtualization can be further improved beyond a single application to encompass a complete workspace, better addressing modern IT requirements. We call this application workspace virtualization or workspace virtualization.

Both technologies are very useful for today's PC world. Moreover, the marriage of the two can allow for order-of-magnitude improvements in the way PCs are managed and used in the future.

In this article, we will examine the features of workspace virtualization, and its applicability to diverse enterprise scenarios such as application deployment, disaster recovery, and creating a managed enterprise "island" on employee-owned and subcontractor-owned PCs

Background
Traditional "desktop" virtualization technology implements a virtualization layer between the operating system and the PC hardware (see Figure 1). This enables several operating systems, each with its associated applications and data, to run on a single piece of physical hardware.

Some desktop virtualization solutions are server-based, enabling many virtual desktop images to run on a single server and accessed from a PC client or thin client through a VDI protocol. The main advantages of server-based desktop virtualization are a reduction in the client hardware costs, centralization of the client OS management, reduction in desktop support costs, and increased data security. However, they suffer from issues of peripheral connectivity (scanners, printers, etc.) and multimedia support (audio, graphics), and require a low-latency connection between the client and the server. Offline functionality is also not available.

Other desktop virtualization solutions are client-based, enabling users to run a separate virtualized operating system on their client PCs. The advantages of client-based desktop virtualization are improved manageability, offline capability, and support for application migration, but it too suffers from several drawbacks: A large footprint for the virtualized image (typically several gigabytes) due to the inclusion of a full operating system image; a considerable performance penalty compared to running applications natively on the host PC due to the virtualization of hardware resources; and the need for an additional (and expensive) OS license for the virtualized environment.

A separate approach to improving manageability and easing deployment of corporate environments is application virtualization (see Figure 2). Application virtualization implements a virtualization layer between an application and the operating system.

The main advantages of application virtualization are solving application compatibility issues, simplifying application deployment and easing OS migration. There are two main drawbacks to this approach: First, most solutions virtualize each application in its own "bubble," isolating it not only from the installed applications on the host, but also from other virtualized applications. The end result prevents virtualized applications from interacting with one another or launching each other. This prevents applications suites such as Microsoft Office, for example, from embedding an Excel worksheet in a Word document or a PowerPoint presentation. Second, most application virtualization solutions require an offline packaging process called "Application Sequencing." This process is partially manual, still considered something of an "art" and does not always end successfully, adding a degree of complexity and uncertainty to the application deployment process.