As data center managers continue to increase heterogeneity to physical environments by introducing myriad virtual environments, there are six major points they should consider.

1.  Application Availability
At the most fundamental level, IT must ensure application availability, regardless of the application or whether it's running on a physical or a virtual platform. To that end, organizations must consider the different kinds of threats to applications and protect against them, including localized component outages on hardware, more widespread outages such as power failure, or even catastrophic site outages resulting from a natural disaster.

Organizations also want to leverage virtualization to provide a higher level of availability for tier two and three applications that haven't been historically included in the organization's high-availability and disaster recovery plans. However, virtual server tools don't protect against application malfunctions caused by the failure of virtual machines or failures in network or storage interfaces. Virtual server tools are often limited to monitoring the physical server failure, which isn't sufficient for production application availability. An application often traverses physical and virtual boundaries, and most solutions only work in a physical environment or in a virtual environment. Such an incomplete solution may provide a false sense of application availability.

2.  Data Availability
As enterprises deploy virtual servers, maintaining an optimal data protection strategy is paramount to ensuring the availability of corporate data. Data in both physical and virtual machines should be protected. Not only should they ensure that data is backed up quickly and without incident, but data should also be easily recoverable.

While many enterprise IT professionals are interested in deploying next-generation data protection technologies in physical or virtual environments, they've been reluctant to do so because of the complexity introduced by numerous disparate point products - each with its own agents, policies, schedulers, recovery processes, user interfaces, and reporting tools. Today, however, new data protection tools are available that provide a single platform that unifies backup and recovery management across physical and virtual environments and offers consolidated backup and granular file-level recovery from a single backup. These capabilities, together with the ability to back up virtual environments to either tape or disk, enhance data protection and availability.

3.  Virtual Server Sprawl
Creating a virtual machine has become an inexpensive and easy answer to so many IT challenges that virtual server sprawl is becoming a worrisome possibility. Some industry experts suggest that virtual server sprawl isn't only inefficient; it causes management challenges.

As enterprises start using virtual platforms in production, they need to ensure that they have a good set of processes and products to control the lifecycle of virtual machines. Virtual machine sprawl can make data center management an arduous task. Complete visibility including discovery and configurations of virtual machines along with the ability to manage a virtual machine's lifecycle is an essential prerequisite to taking it into a production environment.

4.  Physical/Virtual Toolset
Organizations often use one set of tools for managing their physical environment and another for managing their virtual environment. Yet, maintaining such a complex set of processes and tools is inefficient, ineffective, and error-prone.

As virtualization continues to move into production environments, it's imperative to use comprehensive management tools that ensure the availability of critical applications and automate common server management tasks. IT organizations must consider their ability to reuse skills and knowledge to simplify the process of getting virtualization applications up and running in production.

5.  Server Utilization
IT administrators must know how much each server is being utilized, whether additional applications can be put on that server, or if it's running at high capacity.

IT organizations can improve server resource utilization by controlling when and where applications and virtual machines run across heterogeneous environments. IT must be able to track resource utilization and make changes to virtual machines that improve utilization rates and ease overall management.

6.  Vendor Selection
Because moving virtualization from a test to a production environment is both difficult and critical, many organizations may choose to work with vendors that have experience and expertise not only in virtualization technologies but also in data center implementations in complex, heterogeneous environments. Heterogeneity and the resulting complexity, which is the modus operandi of most data centers, should be planned for and controlled. If not, point- and platform-specific solutions will cause heartburn as the data center expands and its management gets exponentially more complicated.

Summary
After years of double-digit IT spending growth, businesses now expect better returns from their IT investments. To that end, they must implement a comprehensive plan for tracking the return on virtual environment investment that measures operational efficiencies, reductions in capital expenditures, and improvements in meeting SLAs.

An old adage says, "If you fail to plan, you plan to fail." This profound statement applies when virtualizing production applications. Thoughtful consideration of management challenges and preparation for a heterogeneous environment will make it easier to succeed in the face of increasing demands from business stakeholders to demonstrate the ROI of your virtualization initiatives.