3 Considerations for Application Performance Management in the Cloud

Your apps may reside in the cloud, but that won't factor into end-user expectations. Experts offer three application performance management monitoring tips to keep in mind.

Esther Shein

June 4, 2012

3 Min Read
Network Computing logo

With more and more companies migrating to the cloud, application performance managementmonitoring is becoming a critical element of the service. Yet few users are adequately prepared to keep an eye on their apps in the cloud, experts warn.

"When parts or all of an application moves to a cloud, the view into the application is disrupted," says Belinda Yung-Rubke, director of field marketing at network performance tools provider Visual Networks Systems. "This may be due to the loss of access to instruments for metrics collection, or access for the data to make its way back to the management system." This creates a disconnect with performance management in the cloud, she says, yet end users expect the same level of availability, access to applications and performance that they receive from non-hosted applications.

Yung-Rube says enterprises must keep the following three application performance management considerations in mind when migrating to the cloud:

1. Enterprises won't move everything to the cloud anytime soon. They will always have hybrid environments, so consider how to manage that."I would expect that the majority of enterprises are going to have cloud and non-cloud environments," she says, but "companies don't think about how inefficient it is to manage disparate tools to manage complex environments." She suggests that organizations consider consolidating tools to potentially cut costs and improve efficiencies.

Greg Schulz, senior advisory consultant to the Server and StorageIO Group, agrees that for the near term and the foreseeable future, private and hybrid clouds will be very popular methods to deliver software, platforms or even infrastructure management. "Some will maintain legacy models while others will leverage a mix of private cloud, legacy and hybrid," he says. "In terms of what they keep internally, that runs the gamut, depending on the organization and their concerns."

"There, of course, will be some that go all public, while others move slowly; some will go all outsource, some managed service, some colo or hosting, and others SaaS/PaaS or leveraging IaaS to different degrees," he adds. "That's the beauty of clouds--you can have and use them your way to meet specific needs in a companion mode versus competitive, where it's seen as having to be all one way or the other."

2. Gain visibility into what the cloud provider is doing."One of the fears our customers have is starting a virtualization project and they can't see anything," says Yung-Rubke. "These days, service providers are not offering service-level agreements because it's too difficult to measure. From the enterprise side, if there are no guarantees it's very difficult for businesses to adopt these things."

It's critical, she says, for service providers with multitenant system services to ensure that data is kept separate. Lastly, she says, they must be able to provide meaningful reports to stakeholders that give any type of information related to the application, including the response time and the amount of traffic.

3. Users will expect availability, performance and response time."They don't really care whether you're going through two service providers or [are] not able to negotiate a service-level agreement--they still expect that, so the end-user experience is still important in a mixed environment."

Schulz, who wrote the book Cloud and Virtual Data Storage Networking, concurs that when it comes to performance management in the cloud, "the top considerations include ability to monitor, track and report on a real-time and historical basis across different providers using their APIs or tools or via third parties." Another factor is the ability to get information and set alerts and triggers for performance (data movement, response time, bandwidth and activity), along with resource usage and availability, including service-level objectives and agreements.

About the Author(s)

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox
More Insights