Archive for April, 2012
Recently I talked with several data center managers about their experiences with virtualization. While these managers have different perspectives, they all agreed that server virtualization alone isn’t enough. By moving beyond server virtualization to a more holistic approach including virtualizing storage, network and other technology assets, these companies are increasing the ROI of their virtualization implementations. There is an element missing in their virtualization strategy that is making it hard to meet the increasing demands of the business. These companies are beginning to reset their virtualization priorities to make sure that all the elements of the IT environment work together – and this includes creating a virtualization environment that automatically allocates resources based on the demands of specific applications and workloads.
IT management needs to focus on the application priorities in terms of performance to support the business. If all your applications are treated with the same priority– how can you be assured that your most critical applications always have access to the resources they require? You may be doing a great job monitoring CPU usage and available memory in your server virtualization environment, but still have unexpected performance problems that impact critical customer applications. What’s missing is a way to adjust for business priority variations when you allocate resources across your virtualized environment.
One way to ensure that your environment operates with an increased awareness of the requirements of each specific application is to implement application infrastructure virtualization. It is a capability that allows for codifying business and technical usage policies to make sure that each of your applications leverages virtual and physical resources in a predictable way. By automating many of the underlying rules and processes, you can optimize both resource utilization and the customer experience at the same time.
There are three main characteristics for application infrastructure virtualization:
- Setting business priorities for applications and automatically adjusting resources to keep customer service levels in balance
- Applying a customer focused approach to the automation of resource optimization so that each application gets the resources it needs based on resource availability and the application’s priority to the business
- Allocating a pooled set of resources to support a group of workloads.
Application infrastructure virtualization ensures that any resource can run any workload. If there are resource limitations then application with the lowest business priority at the time is allocated with the fewest resources.
I amplified this issue in a white paper I recently wrote for IBM on the topic. The paper, called Creating Application Awareness in IT Virtualization Environments, discusses application infrastructure virtualization and how companies can combine server and application infrastructure virtualization to improve overall performance levels. In addition, the paper describes IBM’s solution for application infrastructure virtualization, IBM WebSphere Virtual Enterprise (WVE).
It is easy to assume that server virtualization itself is enough to solve resource management issues in the data center. However, it is increasingly clear that virtualization has to be tied to the performance of various applications based on the times and situations where they demand performance. Tying application performance to virualization creates a more efficient and proactive environment to satisfy customer expectations.