COGNIZANT CONSULTING
Helping organizations engage people and uncover insight from data to shape the products, services and experiences they offer
Learn More
  • Working to reshape business models, modernize products and enhance customer experiences to drive growth.
  • Reinventing and managing your most essential business processes with new ways of working.
  • Simplifying, modernizing and securing the IT infrastructure and applications that are the backbone of your business.
COGNIZANT CONSULTING
Helping organizations engage people and uncover insight from data to shape the products, services and experiences they offer
Learn More

Contact Us

THANKS FOR YOUR INTEREST IN COGNIZANT.

We'll be in touch soon!

x CLOSE

Refer back to this favorites tab during today's session for access to your selections.
Refer back to this favorites tab during today's session for access to your selections.x CLOSE

Perspectives

Monitoring Container Health While Scaling Digital Applications on AWS

2018-11-26


IT container-based technologies help customers materialize microservices architecture, enabling faster and easier development and deployment of services. These technologies are available as standalone or cloud-native services on Amazon Web Services (AWS). Read on for an overview of a few important native and non-native monitoring tools for AWS.

Today, instead of a single virtual machine (VM), there are several independent containerized services, which increase the scope of issues. With production workloads, it’s imperative that there are also adequate monitoring capabilities on these platforms; this is to cover not just the VMs, but also the containers running on top. 

Some challenges to traditional monitoring methods include inefficiency, the complexity involved in 24x7 monitoring of VMs as well as the individual services running within them, categorizing container utilization, and the different layers at which container platforms need to be monitored.

Key Factors to Consider When Choosing Monitoring Tools

With multiple options available, it is important to consider several key factors when choosing an appropriate monitoring tool for the platform hosting the applications. These factors include:

Native and Non-Native Monitoring Tools

In view of these considerations, some important native and non-native monitoring tools for AWS include:

  • Amazon CloudWatch

  • Prometheus

  • Sysdig

  • Datadog

Amazon CloudWatch 

Built specifically for developers, system operators and IT managers, Amazon CloudWatch is the fundamental monitoring service on AWS. To gain deep visibility into ECS clusters, you can augment CloudWatch with a monitoring tool for analyzing performance data at scale. The choice of these tools is dependent on how easily they integrate with CloudWatch. 

Prometheus

Prometheus is an open-source, systems and service monitoring tool that collects metrics (time-series data) from configured targets at given intervals, evaluates rule expressions, displays results and can trigger alerts based on conditions. It has a built-in query language to query a multi-dimensional time series data model, which makes it a uniquely powerful tool. 

Best-Fit Use Case

  • Designed specifically for distributed systems with highly dynamic service-oriented architectures, such as OpenShift, Docker Swarm/Data center, etc.

  • Can be architected for hybrid cloud deployments.

Pros

  • No license cost. 

  • Does not require an extensive infrastructure.

  • Supported by the latest container technologies such as Docker and OpenShift platforms.

  • Leverages open-source tools for integration to PagerDuty, Slack, Jira, etc.

Cons

  • Requires slightly involved setup effort. 

  • Stores data by default only for few days; however, community members are developing adapters for long-term remote storage systems. 

Sysdig

Sysdig monitors AWS Elastic Container Service (ECS) and Elastic Container Service for Kubernetes (EKS) through service-level monitoring of the ECS metadata, making it more dynamic. It allows aggregation of metrics from containers in real time to provide more meaningful views and alerts. Along with Sysdig Monitor, it provides EKS monitoring from a single agent on a platform, while Sysdig Spotlight provides a quick overview of the resources within the EKS cluster.  

Best-Fit Use Case

  • Sysdig can be leveraged to monitor AWS resources along with the container platform and integrates with Docker, CoreOS, Kubernetes, Mesos/DCOS, OCP, AWS ECS, and physical nodes.

  • Sysdig provides a broad range of dashboards classified into platform, technology, host, container level, etc., enabling instant monitoring and creation of alerts, reducing overall setup times.

Pros

  • Per host/pricing.

  • Webhook integration is provided for integration with other tools such as ServiceNow.

  • Guided walkthrough available for installation/deployment of agents for most available container platforms.

Cons

  • Need to install kernel modules on the host machines.

Datadog

Datadog is a full-stack cloud-based monitoring solution uniquely available as SaaS. It provides a single pane of glass for monitoring on-premise cloud and hybrid infrastructure. Datadog application performance monitoring (APM) provides multifaceted performance metrics monitoring and supports integration with ECS and with new metadata endpoints for Fargate. 

Best-Fit Use Case

  • Datadog can be employed for container monitoring along with basic APM.  

Pros

  • Anomaly detection.

  • Containerized agents are available.

  • Ability to create alerts/events from the collected metrics.

  • Datadog has live container monitoring features that let customers monitor running processes, filter and group them using tags, and drill in to see process-level system metrics with two-second granularity.

  • Integration is possible with service management and configuration management tools.

Cons

  • Container metrics are not grouped at container/host/orchestrator level.

  • Not available for on-premise installation, so compliance can be an issue. 

Assess Environments

We have barely scratched the surface of the tools available for container monitoring on AWS. Since available options have their own pros and cons, enterprises may find it prudent to assess each against the environments where it is to be leveraged. Doing so will help select the monitoring tool that’s the most appropriate for the situation. 

Visit us at AWS re:Invent at booth #1001.

For more information, visit the Cognizant Digital Systems & Technology section of our website, or contact us.

This Perspectives piece was written by Narsi Balasubramanian, Chief Architect and head of the Center of Excellence for Public Cloud and Containers & Platforms, Cognizant. He can be reached at Narasimhan.Balasubramanian@cognizant.com.

Related Thinking

Save this article to your folders


Save

PERSPECTIVES

Powering the Enterprise with...

As organizations move forward with microservices, they need to be...

Save View

Save this article to your folders


Save

PERSPECTIVES

How Cognitive Computing Unlocks Business...

Extending process management to business logic offers enterprises...

Save View

Save this article to your folders


Save

PERSPECTIVES

Simplify, Modernize and Secure Your...

What does it take to provide digital products and services? Everything...

Save View
Monitoring Container Health While Scaling Digital Applications on AWS