For the last decade IT managers have been transitioning from largely on-premises infrastructures to cloud environments such as Amazon Web Services (AWS) that are easier and less costly to deploy, manage and scale. At this point cloud platforms like AWS are often the default deployment preference, and IT managers have new worries about managing cloud performance and cost. Cloud performance monitoring is key to taking full advantage of all of the advantages that AWS can offer.
When Clouds Lead to Alert Storms
The nightmare scenario. The usage of a mission-critical cloud service hosted on Amazon ECS (Elastic Cloud Services) has pegged for unknown reasons. Now it’s throwing errors and causing a cascade of failures, both in the EC2 environment and across applications hosted in-house that reach out to the service. Amazon provides some default monitoring, but your custom apps and virtual servers are complex and interdependent. Alerts are coming in fast and furious from all corners, and irate texts from frustrated execs are blowing up your cell.
To avoid the nightmare scenario it is best practice to develop an application and network monitoring strategy that covers your entire infrastructure. Your strategy defines which applications and services are most critical for your business, how you are going to monitor them, what network monitoring tools you will use, and what should happen when incidents arise. What’s most important is that your monitoring strategy equally embraces private network and public cloud infrastructures. Pinpointing the cause of an issue can be devilishly hard without a hybrid cloud monitoring tool that straddles both your own network, Amazon Cloud Services and any other critical cloud-based networks or applications.
The Costs of Cloud Computing
One big reason that IT managers embrace Amazon Cloud Services is that doing so can lower overall infrastructure costs and make them more predictable. But do you know if you are using all of the capacity you’re paying for? Unlike your physical network environment, cloud services cost more when they are used more. To “right size” your investment in AWS, It’s best practice to use a cloud monitoring solution to stay on top of cloud performance metrics such as uptime, CPU utilization, network utilization, disk performance, disk reads/writes as well as memory utilization, disk swap utilization, disk space utilization, page file utilization, and log collection.
Using the data from the Amazon AWS API, your cloud monitoring tool should be able to send proactive alerts so that you are never surprised by large swings in your monthly bills.
Trust But Verify
When IT managers contract with cloud providers, they sign a contractual guarantee of performance, or SLA (service level agreement). If Amazon doesn’t uphold their end of the bargain and deliver the uptime promised, organizations get service credits in compensation. With a robust AWS cloud monitoring tool you can verify the performance that Amazon is reporting.
Monitor Everything in Your Network
Start Your Free Trial of WhatsUp Gold