Fantastic 4: Must-Have Visualizations for Your Cloud Monitoring

November 5, 2018 Gaanesh Kapatralla

The Wavefront team has released a wave of new dashboard features including new visualizations, interactions, and updates. These new features help DevOps and SREs to monitor and troubleshoot a wide range of application and infrastructure use cases. In this blog post, I’ll share the inspiration for these new dashboard features, walk you through an example, and show you how to get started.

Charts for Holistic Cloud Monitoring

At VMworld 2018, an SRE at one of our customers made it clear to me why he needed a set of interlinked dashboards with a comprehensive multi-level view of the entire cloud infrastructure when he said:

“It’ll make my life a lot easier if I get infrastructure dashboards that show me the health of my overall AWS environment at various levels of aggregation, and that indicate the trouble spots. Currently, it takes me hours to monitor each instance individually, which can be a quite frustrating experience.”

Wavefront reduces AWS monitoring hours to seconds by introducing globe charts, node map charts, top K charts, and status list charts.

Globe Chart (Figure 1) helps you visualize the health of your infrastructure located across the world. You can set up the globe chart to get a 30K feet view of the overall health of your infrastructure. You could even set up this chart in the office lobby to display your company’s worldwide sales, or in the war room to monitor your app uptime!

Figure 1: Globe Chart

Node Map Chart (Figure 2) gives a regional overview of your EC2 instances so that you can group, filter, and explore all your instances on the fly. Each node is color-coded based on the status of the instances to help you quickly visualize and troubleshoot infrastructure health problems across the world. Fundamentally, a node map chart enables you to visually track minimal changes among several homogeneous objects at a glance. You can select different metrics and group the nodes based on tags, or hover over a node to display a preview of the line chart for the selected instance.

Figure 2: Node Map Chart

Top K Chart (Figure 3) displays a sorted list, in descending order, of the instances and their corresponding values in the specific region. The length of the bar shows how the value for each instance stacks up relative to the values of other instances. Hovering over a bar displays all the associated tags.

Top K charts help you quickly identify the instance with the highest resource consumption, such as CPU or network ingestion, and act accordingly.

Figure 3: Top K Charts

Status List Chart (Figure 4) displays the status of your AWS infrastructure, aggregated by region, in different colors and icons that indicate severity. Engineers can map icons and colors to show different severity levels visually.

Figure 4: Status List Chart

Dashboards Interactions for Magic Drill-Downs

Some of the other new features are the revamped preset date and time ranges, breadcrumb links, and dropdown variables. With dashboard linking, you can connect multiple dashboards and form a drill-path for guided analysis. With row and column span your charts can span across multiple rows and multiple columns in a dashboard.

Let me illustrate you the benefit of dashboard linking magic with an example of Jason, a DevOps Manager for a popular peer-to-peer ride-sharing app used by millions of users daily. He uses Wavefront to monitor the company’s AWS infrastructure and ensure 24/7 up-time for the ride-sharing app. It’s New Year’s Eve, and increased app usage causes multiple server outages in the us-west-2 region, which is shown as red on the globe chart as well as on the status list (shown in Figure 5).

Figure 5: Global Dashboard

Jason clicks on the red us-west-2 and drills through to the regional dashboard (shown in Figure 6) where he notices a red cube, which are EC2 instances.

Figure 6: Regional Dashboard

He clicks on the red node to drill through to the instance page (shown in Figure 6) where he figures out that the instance is over-utilized. Here’s the problem!

Figure 7: Instance Dashboard

Because Jason knows that higher provisioning will solve the problem, he requests a higher CPU resource limit from AWS and is relieved that he found the root cause of the problem so fast.

Ready to Get Started?

If you are an existing Wavefront customer, follow these steps to try the new AWS EC2 read-only dashboards:

  1. Log in and select Browse > Integrations.
  2. Select the AWS integration.
  3. Scroll down and select the Dashboards tab. You will be redirected to the AWS dashboard page, where you will find a link to the new dashboards on the top right.
  4. Click on the Preview brand new Amazon EC2 dashboards link (see Figure 8).
  5. Et voilà! Start experiencing the benefits of the new AWS dashboard features.

Figure 8: Access to the new dashboards

For more information please contact Wavefront team or slack us at Wavefront public slack.

Interested in learning more about Wavefront? Check out the Wavefront free trial.

Get Started with Wavefront Follow @Gaanesh_K Follow @WavefrontHQ

Co-authors:

Durren Shen
Durren is a co-founder of Wavefront. He previously worked on start-up products including a trading platform for money market instruments and an ELO algorithm based product used to rank online contests. He likes to get many types of interesting data into Wavefront for analysis in his spare time.

Jiwoong Youn
Jiwoong is a Member of Technical Staff at VMware, working as a front-end engineer in the Wavefront team. He majored in Computer Science and graduated from the University of Illinois at Urbana-Champaign. Prior to VMware, he worked at AppDynamics as a front end engineer working on data visualizations.

Jason Chu
Jason is a Staff Engineer for the Wavefront Front-end team. He found himself fortunate having worked on a diversity of web front-end projects in different industries in Silicon Valley. He is passionate about user experience and enjoys a great deal implementing time series and other chart types in Wavefront.

Abhilash Srivastava
Abhilash is a Front End Engineer at Wavefront, by VMware. He enjoys Web development and all things Javascript.

The post Fantastic 4: Must-Have Visualizations for Your Cloud Monitoring appeared first on Wavefront by VMware.

Previous
One Billion Lyft Rides, Half Million Smart Homes: How Centrica Scales SaaS with Analytics Insights (Part 1 of 2)
One Billion Lyft Rides, Half Million Smart Homes: How Centrica Scales SaaS with Analytics Insights (Part 1 of 2)

The post One Billion Lyft Rides, Half Million Smart Homes: How Centrica Scales SaaS with Analytics Insights...

Next
5 Ways to Save on Your Google Cloud Services Bill
5 Ways to Save on Your Google Cloud Services Bill

Google Cloud Platform (GCP) is a “top three” public-cloud computing provider along with AWS and Microsoft A...

Visionary in Gartner® Magic Quadrant™

Learn More