– Zohar Gilad, executive vice president of products, marketing, and channels at Precise (www.precise.com), says:
When it comes to application performance, data center managers don’t usually dig too deeply. They are embroiled in the day-to-day requirements of managing the infrastructure that sits beneath the applications. Yet when users start complaining about slow response times or application errors, the IT managers who oversee those applications come knocking at the data center door with questions. How is the network/storage/servers affecting my system? Data center managers don’t always have those answers at hand. They may not have the tools and resources to monitor the infrastructure in the context of individual applications, particularly in environments with hundreds of servers and dozens of applications. They can share network statistics and CPU utilization rates, but they often can’t tell the application guys how to solve their problem or even where to begin troubleshooting.
And then along came the cloud. Virtualization technologies and private clouds have made the tracking and management of virtual infrastructure resources easier for data center managers. Virtualization software has self-management capabilities that automatically take care of a resources-starved virtual machine and move it to another server.
That visibility is a distinct benefit for the data center staff. They can now view the virtualization console for comprehensive information about IT assets, where servers and VMs reside on the network, and their vital statistics. Yet at the same time, monitoring applications in the cloud gets harder. Traditional systems monitoring tools that data center managers use aren’t as effective in the cloud, when performance begins to suffer. Those tools are typically point systems for monitoring specific pieces of the infrastructure, such as storage or servers; they don’t typically contain application transaction-level data. These monitoring systems are also not designed to analyze the fluid infrastructure of the cloud.
Virtualization and cloud technologies create unique challenges for applications that need to be adjusted for dynamic provisioning: How applications start and connect to servers will change as the virtual environment allocates and shifts resources based on policies and hardware availability. For instance, some applications have hard-coded server names: this doesn’t work in the cloud. Data center managers need to address the following cloud-specific infrastructure issues that impact applications:
• Degradation of service during the phases of transition from physical to virtual infrastructure (and thereafter);
• The potential negative impact on quality of service (QoS) of inter-application shared resource contention;
• Degradation of QoS at peak application load times.
Transaction Performance Management or TPM software can help measure and define QoS in the cloud by recording how transactions are performing and what components of the infrastructure they rely upon. With TPM, an organization can associate transactions to users and trace transaction paths through both virtual machines and physical servers, in real-time. As just one feature of a cloud-ready TPM system, consider this scenario: A virtual machine that’s on your network in the morning may be decommissioned in the afternoon, and all of the relevant monitoring data within it also disappears. However, a TPM system retains that performance data so that you can go back and understand what happened and where during a transaction window, to determine the source of the problem.
IT departments can purchase TPM as a virtual appliance that is downloaded and provisioned in less than 30 minutes. Companies can deploy TPM just like any other virtual machine and have it immediately available to monitor critical business applications. It’s time for data center managers to get out of the data center, and get closer to the end-user applications that run the business. Virtual and cloud technologies, with the right monitoring software, help ease that transition for the data center manager.
