alphaspirit - Fotolia

Manage Learn to apply best practices and optimize your operations.

Strict monitoring keeps mission-critical apps out of trouble

Mission-critical apps are bound to get into trouble if you take your eyes off of them. Being strict with performance monitoring can ensure the best availability and security for your vital applications.

Mission-critical applications must be monitored continuously so that any disruption in availability or performance can be addressed immediately. And despite some fears, moving applications to public or hybrid cloud vendors doesn't mean you miss out on the opportunity to keep an eye on your apps. In fact, close monitoring must also carry through to mission-critical apps relegated to the cloud, including checking in on overall availability, relative performance (such as transactions over time) and the amount of resources used by the workload, which will affect its monthly hosting costs. It's this monitoring that can ensure the safety and performance of these vital applications.

Cloud providers rarely provide users with more than a glimpse into their infrastructure. For example, many services supply a Web-accessible dashboard or other view that reports aggregate cloud resource usage, such as CPU, memory and disk use. Cloud services geared toward important workloads may support more established and sophisticated workload management tools though agents or application program interfaces. As an example, VMware vCloud Air services support other VMware tools such as vCloud Automation Center, vCloud Connector, vCenter Orchestrator and vCenter Configuration Manager, along with third-party tools such as Hyperic HQ.

Cloud providers rarely provide users with more than a glimpse into their infrastructure.

Regardless of the actual management and monitoring tools, it's worth a prolonged proof-of-principle project to test the workload and verify its interoperability with prevailing tools or platforms -- know that comprehensive monitoring and management are available and will actually provide necessary details to local administrators well in advance of general workload deployment to the cloud. Understand how to watch performance trends, change resource allocations and react to availability problems through the provider's support mechanism. If this doesn't meet your expectations in a proof-of-principle project, it probably won't get any better putting the vital workload in production. In that case, the best option may be to keep running your application in-house.

About the author:
Stephen J. Bigelow is the senior technology editor of the Data Center and Virtualization Media Group. He can be reached at

Next Steps

Netflix an A student for cloud app availability

IT pros misinformed on risks of cloud computing

Dig Deeper on Cloud infrastructure monitoring

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.