10 Best Practices for Effective Application Management in a Virtual Environment

Date:   Wednesday , November 03, 2010

Virtualization has changed the world of IT, as it provides significant cost savings and business flexibility. It also brings new and difficult challenges to application management, making the traditional, physical performance monitoring tools and approaches insufficient for the virtual world.

Virtualization slows down resources and may create conflicts at the physical level that will impact application performance. Other challenges facing IT include monitoring the health of the virtual machines, analyzing data, alerting staff to problems, and performing additional administrative tasks. All of these issues require both a new approach and new generation of solutions to ensure continuous service delivery and support.

Virtualization facilitates increased business continuity, disaster recovery, flexibility, and agility; reduced downtime and cost, operational efficiencies in backup, high availability and storage. Savings come from the consolidation of hardware, including maintenance, hardware upgrades and additional application-driven hardware purchases that no longer are necessary, as well as elimination of multiple software purchases for all the hardware. Reduced man power and less power consumption also means reduced utility costs.

Many organizations don’t fully reap the benefits of virtualization because their virtual infrastructures are hindered not just by the challenges, but also by poorly performing applications that don’t meet SLAs or end-user needs. End users need to stay productive and satisfied with the performance of the system, so an effective application management plan should, first and foremost, consider their perspective and eliminate performance issues that hinder their productivity.

Good planning and a set of best practices are the keys to a well-developed virtual environment that achieves optimal performance, contains costs and provides significant ROI to the organization, and ensures the virtualization initiative effectively supports end-user objectives and business-critical applications.

To meet the operational challenges of virtualization you need to have a thorough understanding of your virtual infrastructure, and manage the relationships and interactions between all of the components in the environment. To start with, thoroughly understand the technological maturity of your organization. Does it have a low level of maturity, with little monitoring, a chaotic environment and frequent customer complaints? Or, is it a highly matured organization in which SLAs are always met and customers are totally happy.

An application performance management plan must be a continuous improvement process that evolves as the needs of the organization change. It is important to continuously understand and monitor your organization’s maturity at all levels once initiated. Other steps to mitigate performance issues that affect business-critical operations include measuring everything continuously, choosing software solutions judiciously and introducing and embracing new processes. This includes restructuring the service model to more of a service management model that focuses on the end user’s perspective of the quality of IT service.

What is a best practice?
Application performance management is about people, process and technology bound together to keep the business competitive and attractive, and to retain customers by improving the output to the end users.

A practice becomes “best” when continual measuring of the process adjustment indicates you have moved from where you started to where you wanted to go. Following are 10 best practices for effective application management as you begin building your virtualized environment:

1. Develop an application performance management plan
The plan is the most important part of application performance management and should be undertaken from the end users’ point of view. Understand end users’ expectations, develop SLAs that are meaningful to them, and recognize that you will continually tweak your plan as more of your environment is virtualized.

2. Create a project team that has the same goals
The team must understand that for a virtualization initiative to be successful, the organization must walk before it can run. This means starting small, measuring and monitoring the parts of your environment that have been virtualized, and slowly adding more. It means adjusting to new tools and approaches to effectively protect and maintain the environment.

3. Make sure the end user perspective and your applications’ capabilities in the virtual environment are communicated, documented and understood throughout all of IT
This is critical to improving the delivery of performance to end users, as well as making sure IT is aligned with the business goals.

4. Understand all components of your environment and how they interact
Many of the challenges of managing applications in a virtual environment stem from the way virtualization technology leverages the four core resources: CPU, memory, disk, and network. By understanding the impact of resource sharing, and gaining insight into the key components of the virtual environment – data centers, data stores, clusters, resource pools, ESX servers, and VMs – your organization can better support the applications running in the virtual environment.

5. Know your applications and what virtualization will do to them
Know the application’s technical parameters, measure and monitor them continuously before you start building your virtual environment and after.

6. Carefully choose the software solutions you need to facilitate optimal performance of your virtual infrastructure
Weigh the complexity and the cost of solutions, and choose those that allow you to see the entire virtualization infrastructure in a single view to gain understanding of how it impacts the entire application environment.

7. Have the capability to determine the root cause of an incident or problem before end users are affected
A good understanding of the way the components of your virtual environment work and interact will help to determine the root cause of incidents, and meet performance and SLA objectives. Combine that understanding with tools that help you determine a problem exists, communicate why it’s a problem and show how to resolve it
.
8. Measure business-critical application performance and communicate findings to business leaders
Show them the business-critical application performance history in a way that is relevant to them (number of successful transactions, availability, etc.).

9. Act on your findings
With business leaders, examine the application performance history to determine what the problems are, and create a plan to improve it. Make sure you take into account both current and planned future work, and any changes in business priorities. Reprioritize your processes from both IT and business perspectives, and make the changes that will provide the most return and flexibility, but also reduce cost and risk.

10. Continuously measure and monitor everything in your virtual environment
Start by measuring what you understand, with the end user in mind. Track measurements over time to see trends, and develop a baseline. You need to know where you are to show improvement when your processes take you to a different level.

Virtualization must be implemented with care and planning if it is to successfully support applications and minimize negative effects on end users. With a sound plan, processes that promote optimal performance of the virtual infrastructure, knowledge of how and where to contain the costs of implementing a virtual environment, and a good understanding of both the environment and its applications, you can virtualize with confidence and your organization can fully realize the benefits of virtualization.

The Author is Head - ADM, Quest Software