Best Practices

Microsoft Azure Cloud Monitoring

Azure cloud monitoring should be one part of a holistic network and application monitoring strategy.

Microsoft Azure is currently the fastest growing cloud infrastructure provider, with revenues growing 76% year over year in 2018, according to a June 2018 earnings release. The platform competes with Amazon Web Services (AWS) and is not just for Windows applications. Azure can host Java, Python, Ruby and Node.JS apps as well as storage, databases, virtual machines and anything else that IT managers might once have hosted in-house. However, as with any cloud hosting environment, IT managers should have a holistic cloud monitoring strategy in place before migrating mission-critical apps to Azure.

Monitoring, Not Just Cloud Monitoring

Most organizations do not host 100% of their IT infrastructure in the cloud. IT Managers need to keep their fingers on the pulse of complex, multi-vendor environments with inter-dependent services both in the cloud and on-premises. When one or more of these services are hit by an outage, it’s important to be able to quickly pinpoint and resolve the issue, wherever it arises, before users are impacted. Microsoft Azure provides full-stack cloud monitoring tools, but it is best practice to make troubleshooting easier by having a single, integrated dashboard that monitors the state of your organization’s entire networked environment, both cloud and on-premises.

How to Discover Azure Virtual Machines in WhatsUp Gold 2018

A Template for a Hybrid Monitoring Strategy

To develop an application and network monitoring strategy, you first have to decide what you have to monitor. Does your existing monitoring tool enable you to monitor the basics like availability/uptime, disk capacity and response time, memory utilization and log collection across Microsoft Azure and your in-house networks? What are the key applications and baseline services that are most important to your business? What are the risks when there is an outage? Can they go down for a day? An hour? Just a few minutes? Or never? Where do these applications and services live? Often large organizations have thousands of cloud apps and environments to monitor and manage.

Once you have determined WHAT metrics you are going to monitor, you have to understand HOW you are going to monitor them and what should happen when incidents arise. Your network and cloud monitoring tool should have customizable alerts and flexible thresholds to ensure that the right person (or automated process) is alerted/triggered at the right time.

Trust but Verify

The main reason that IT teams turn to cloud platforms like Microsoft Azure is to have predictable and scalable environments with lower overhead costs and risk than self-hosting. So it’s very important to have a cloud monitoring tool that tracks the metrics that prove the return on investment for Azure cloud deployments. There are specific Microsoft Azure cost metrics to monitor like: CPU credits remaining and consumed, resource utilization, overages and burn rate. A good Azure monitoring tool will ensure that you are never surprised by your monthly bill.

One best practice is to look at each of the SLAs promised by Microsoft for your Azure environment and make sure that you have an independent monitor in place for each. For instance, Microsoft guarantees a 99.9% response rate for API Management Service instances. Using a third-party network monitoring tool you can easily verify that the uptime reported by Microsoft is matched by your own Azure cloud monitoring tool.

Monitor Everything in Your Network

Start Your Free Trial of WhatsUp Gold

  Download Free Trial     Watch a Demo