By Bo Barker, Vice President of Services
The amount of people with a stake in your IT infrastructure performance is significant. It’s not just you and your team. Its end users, customers, co-workers — anyone and everyone who relies on the applications you support. None of these people really care about capacity planning, technology upgrades and refreshes, or any other concept that leads to performance degradations and unplanned outages. They just want their applications to be highly available and highly performant all the time. Delivering on these expectations calls for a proactive approach to performance management. Here are three things you and your teams need to know to meet those expectations, and to solve the most challenging problems in infrastructure performance management:
Know what your end users expect
Your end users have expectations. Simply put, they want applications to work. It’s like oxygen in the atmosphere – they expect it to be there. Understanding what it takes for your IT infrastructure to support and deliver mission-critical workloads will help you establish accurate baselines so you can guarantee availability and performance. You need to continually measure workloads in order to tune your infrastructure to deliver on the service objectives your applications require. Quick fixes and reactionary measures or, worse, spending increasing amounts of money overprovisioning are all impractical in today’s IT. Once you’ve established baselines through accurate measurement and models for performance, you can make certain the systems are in place to keep your end users happy.
Know what has already happened to prepare for what could soon happen
As the decades-old saying goes, “Those who fail to learn from history are doomed to repeat it.” Your infrastructure generates massive amounts of performance data every second. Aggregating, correlating and analyzing all of this information, and transforming it into answers that drive decisions, helps you and your teams proactively manage and optimize performance. This real-time monitoring and analysis also helps you avoid predictable issues, and minimize (and often eliminate) predictable risks. Understanding seasonal trends that lead to significant increases in traffic and activity must inform decisions related to planning, tuning and upgrading. Similarly, when you discover the root cause of a recurring problem, it’s not enough just to monitor until the problem is fixed, you must monitor continuously in order to ensure the problem is eliminated altogether. You don’t have time – and your end users don’t have the patience – to wait to solve problems that you could have avoided.
Know you’re looking for answers and nothing else
You have relationships with your vendors. You also have established processes and management methodologies. Ultimately, neither of those things matter in solving performance management problems. You need solutions and answers to your questions. Analyzing complete workflows and the breadth and depth of data from your IT infrastructure activity will give you the answers you need – so long as your monitoring and analysis solution can handle the continuous, real-time nature of performance data and leverage an unbiased approach without favor toward any metric, concept or vendor. If you identify the answer to a problem, implement it, and rest assured that you have the data you need to prove and support your decisions.
Changing the way your team approaches performance management starts with understanding what you need. A critical infrastructure audit from Virtual Instruments will get you started.