Container adoption has spiked in the last few years. Customers are embracing docker containers to leverage new levels of flexibility and agility. Gartner reports that “By 2018, more than 50% of new workloads will be deployed into containers in at least one stage of the application lifecycle.”1
We at CloudHealth Technologies have been helping our customers with their AWS, Azure and Data Center environments for a long time. Over the last year, we worked closely with some of our container customers to help them address their container governance and optimization challenges. To help solve these challenges and to promote best practices learned during our research, we recently launched the CloudHealth Container Governance Module. This new offering, which gives customers visibility into their Kubernetes and Mesos environments, and help identify areas for optimization, is already generating significant excitement among our customers. Don’t believe me? Read what Ben Chess at Yelp has to say. “CloudHealth lets us maximize the value of our container deployment by telling us how well utilized it is and whether our clusters have the right mix of resources supporting it. That level of insight enables us to make informed, strategic business decisions without additional overhead.”
Let's look into some common set of challenges that customers face around containers and how CloudHealth can help solve them.
Problem: “I had no way to determine which services are consuming my Kubernetes cluster resources” said an IT ops manager at an online retailer. “Can you help me with that?”
Solution: The CloudHealth Container Module helped the IT Ops Manager gain granular visibility into the CPU and memory consumption by individual service, or groups of pods, in their environment. With that information, the IT Manager can track and trend resource consumption by teams and services or over time (see diagram below). Using this data now he wants to introduce showback model to different departments and lines of business.
Problem: “We spent a week trying to decide which instance types, and how many of them we should deploy to support our OpenShift clusters. And since the environment constantly changes, I’m not sure these are the right resources to support the cluster anymore” said the director of software engineering at a non-profit organization.
Solution: The director of software engineering can visualize his cluster resources broken down by EC2 instance type, and identify where they have memory heavy resources deployed used for memory-light use cases. Using CloudHealth’s platform they can see if right resources are allocated at the right location. The following is a report from CloudHealth platform that helped the customer identify that they should consider changing their r4.4xlarge instances to a smaller type, or a different family due to low memory utilization (r4 is a memory-intensive family type).
Problem: A Business Analyst from a multinational financial services firm said “People provision much more than they need and more than a third of the AWS fleet supporting our cluster is underutilized.” He continued “we have a historic lack of cost accountability which is a big concern”.
Solution: While working with this customer we found that they had one small group who managed the cluster fabric, but hundreds of other teams were consuming resources on the cluster. With CloudHealth, they could see the relative usage of the different teams of the cluster resources and identify areas where teams were requesting more resources than they truly needed.
In summary, CloudHealth can help you provide the insight to optimize your container environment from cost, usage and reporting perspective.
Want to learn more about our new container module? Get in touch to see a demo of our platform in action in less than 20 minutes. In the demo, we will walk you through the rich feature set and how you can gain complete visibility of your environment through simple clicks.
1Gartner: Containers Will Change Your Data Center Infrastructure and Operations Strategy, March 2016