Hybrid cloud adoption more than tripled during 2017. As enterprises move more of their resources to cloud infrastructure, the need for physical infrastructure monitoring begins to diminish. However, there are a number of good reasons from a security and architecture perspective for keeping some resources internal, especially during a transition period. Even if the goal is to move all IT resources to the cloud, that change doesn’t happen overnight and the challenges of managing this hybrid environment would still exist. How else can user service levels and operational efficiencies be maintained sans this visibility even if control of the physical infrastructure is outsourced?
Many cloud providers offer their own monitoring and reporting services. Some are rudimentary while others are more advanced. Unfortunately, this scenario creates multiple points of administration, potential data-sharing problems, and necessitates training users in multiple tools. If this narrative sounds familiar that’s because it is. However, instead of having an “in-house” silo-specific problem it’s been shifted to the cloud services provider. It doesn’t matter where the silo is it’s still redundant and wasteful.
It’s a well-known tenet of enterprise management that the fewer places you need to look for
information, the faster encountered problems are identified and resolved. Translation: IT teams are more effective when a consolidated approach is employed. Visualizing and managing local and cloud hosted platforms in the same toolset should be the goal. However, cloud providers don’t always offer explicit availability guarantees for application-level functionality. Therefore, the burden falls on the l&O team to choose applications that can handle both environments well.
Hosted virtual machines and storage do offer operational redundancy. However, crafting truly resilient application architectures within the cloud can be tricky, and can still carry the risk of degraded user performance once complete.
The key to identifying and tackling this challenge is integrating application performance monitoring combined with a resource-focused monitoring tool. Ideally, your application monitoring platform supports an integrated global view of all your enterprise resources regardless of their location. The ability to monitor the performance of applications from various locations around the Internet, as well as from where it is hosted is paramount. Only then can you ensure the metrics shown give you an infrastructure as well as a “user-experience” perspective into your application. The best of these tools can integrate all of that functionality and visibility directly into a single console.
Many cloud providers like Amazon AWS have advanced tools for controlling the security and traffic flows within their cloud, but these highly granular controls are complex to configure and create the potential for misconfiguration, which can cause degradation and service interruption.
All organizations must have a plan to ensure reliable application delivery with rapid notification of performance issues and failure. An effective application monitoring platform must provide detailed validation of content, including the ability to look for XPath or regular expressions to fully validate content and guarantee that applications are performing as required.
Additional key functionality is the ability to integrate with a cloud provider’s APIs to automatically detect new services, consolidate events, and alerts and logs. When this capability is present the time and effort saved monitoring, administering, and on-boarding new resources is incalculable. In contrast, sans this capability, a monitoring platform that fails to adjust dynamically has more dire consequences: unreported/undetected user-facing errors and rapidly degrading user confidence in services – and much bigger problem to rectify.
While the move to a hybrid cloud architecture is inevitable for most organizations, replete with its many advantages, a consolidated monitoring approach will help to mitigate the inherent performance risks that can occur if care is not taken from the outset.