What are Containers?

Containers have long been used in the transportation industry. Cranes pick up containers and shift them onto trucks and ships for transportation. Container technology is handled in a similar vein in the IT world. A container is a new and efficient way of deploying applications.

A container is a lightweight unit of software that includes application code and all its dependencies such as binary code, libraries, and configuration files for easy deployment across different computing environments. Since a container is self-contained and includes all dependencies, the application it supports will run reliably across different computing environments.

Containers are used to get software to run reliably when moved from one computing environment to another (e.g., from a physical machine in a datacenter to a virtual machine (VM) in the public cloud).

The term “container image” refers to the package of software that includes everything needed to run an application: code, runtime, system tools, system libraries and settings. Container images become containers at runtime. Containers execute on a Container engine. There are different container engines available today. The popular ones include:

  • Docker: A popular open-source platform, Docker uses the capabilities of a Linux Kernel.
  • CRI-O: A lightweight open-source container engine created by Red Hat. It is the first implementation of CRI (Container Runtime Interface) and offers an alternative to Docker. It is widely used as the runtime engine for Kubernetes (K8s), a popular container orchestration system by Google. Kubernetes can use any OCI (Open Container Initiative) compliant runtime with CRI-O.
  • Containerd: A daemon for Windows and Linux from the Cloud Native Computing Foundation. Containerd can manage the entire container technology lifecycle, starting from image transfer to container execution and beyond. The containerd plugin cri allows developers to use it as the container runtime for Kubernetes.

Do You Need Containers?

Containers might sound like new, flashy technology, and you may wonder why you would ever need them. Yes, we need them. We have all been in a situation where we tried to run an application and failed because it had a missing dependency or had a dependency that was incompatible with the environment in which it was being executed, such as an older version of Java or PHP.

What is the Difference between Containers and Virtual Machines?

With virtualization, the unit of packaging and delivery is a virtual machine (VM), which includes an entire operating system (OS) along with the application(s). In contrast, a container includes an application and all of its dependencies. A container shares the operating system kernel with other containers executing on the same node. This means the containers are far more lightweight, are faster to spin up and use far fewer resources than virtual machines. For a more detailed analysis of the differences between containers and VMs, see: Containers vs VM & Virtual Machines | eG Innovations.

Figure 1: Comparing Virtual Machines and Containers

A virtual machine may take several minutes to boot as its operating system has to start up. On the other hand, containerized applications can be started almost instantly. That means containers can be instantiated in a “just in time” fashion as they are needed and can disappear / be deleted when they are no longer required, freeing up resources on their hosts. So, containers offer a level of dynamicity that virtual machines cannot support.

What are the Key Benefits of Containers?

Containers offer a practical way to build, test, deploy, and redeploy applications on multiple computing environments. Their benefits include:

  • Lower overhead
    Containers require less system resources than virtual machines because they don’t include entire operating system images.
  • Enhanced portability
    Applications running in containers can be deployed easily across computing environments – on-prem and cloud, different container engines, etc.
  • Ensure consistent operation
    Applications deployed using containers will run the same, regardless of where they are deployed, so there are fewer surprises in production environments. Eliminates the “but it worked on MY computer” factor when reproducing bugs and support issues.
  • Greater efficiency
    Containers allow applications to be more rapidly deployed, patched, and scaled.
  • Streamlined application development
    Use of containers can accelerate development, test, and production cycles.

Containers are a natural choice if you’re looking for a technology that will make your life easier and simplify how you deploy and operate applications.

What is the Relation between Containers and Microservices?

Containerization allows for greater modularity in application design and deployment. Instead of running an entire monolithic application inside a VM, the application can be split into components or modules (e.g., the front-end, the middleware, the database, and so on). This is the microservices approach. Applications built in this way are easier to manage because each module is relatively simple, and changes can be made to modules without having to rebuild the entire application. Because containers are lightweight, individual modules (or microservices) can be instantiated only when they are needed and are available almost immediately. Furthermore, the number of containers required for each module can be dynamically determined and scaled based on the processing requirements of that module. A microservice architecture can also offer security benefits too.

What is Kubernetes?

Containerized applications can get complicated. Since containers are easy to spin up, there could be hundreds of separate containers deployed in production. Therefore, there is a need for tools to orchestrate or manage all the containers in operation. One of the most popular container orchestration tools is Kubernetes (sometimes known as “K8s”).

Kubernetes orchestrates the operation of multiple containers. It manages the assignment of resources from the underlying infrastructure to containers – this includes compute, network and storage resources. Orchestration tools like Kubernetes are a must if you have to automate and scale container-based workloads in production environments.

Is there a Standard Format for Containers?

The main standards around containers are:

How is Container Monitoring Done?

Container monitoring is the process of tracking the operation of a containerized environments. Since containers are ephemeral in nature, they are more difficult to monitor compared to traditional applications running on virtual servers or physical servers. At the same time, container monitoring is an important capability needed for applications built on microservices architectures to ensure optimal performance.

Don’t use Legacy Monitoring Tools for Containers and Kubernetes Monitoring

“In an orchestration system such as Kubernetes, you will cause yourself great pain if you try to treat it like your legacy data center. In a legacy data center, everything is fairly static, but in Kubernetes, almost nothing is static.”

https://techbeacon.com/enterprise-it/top-container-monitoring-challenges-how-overcome-them

Since containers can be short-lived, container technology introduces several challenges for monitoring tools:

  • Since a container’s lifespan could be as small as a few seconds, tracking events regarding container activity is very important (polling cannot always be relied upon because containers can be created and destroyed in the polling cycle, i.e., between successive polls).
  • Since containers are lightweight, it may not be advisable or possible to deploy monitoring agents inside each container.
  • Traditional monitoring tools that analyze data flows by tapping into network switches often cannot be used, because communication can happen between containers on the same node.
  • Since containers are short lived, any monitoring instrumentation that is required has to be enabled automatically – i.e., deployment of the monitoring instrumentation has to be integrated with the lifecycle of the container.
  • From an analysis perspective too, there are changes necessary. There could be hundreds of containers that have been started and destroyed during a day. To see the overall performance, monitoring and reporting tools must allow IT admins to look at performance metrics across all the containers (including ones that may not be operational when the analysis/report is being taken).

Container monitoring tools are required to track the state, resource usage and traffic generated by each container. At the same time, it is important to track the performance of the nodes that containers run on. After all, if there is a bottleneck on a node, it impacts the performance of all the containers running on it. Monitoring tools often go beyond just monitoring nodes and containers. They also track the health of applications running on the containers (e.g., tracing the health and performance of web transactions being handled by the Java application server running on a container).

eG Innovations has a variety of capabilities to support monitoring of containers and Kubernetes and the applications they support.

Container Monitoring Dashboard

Figure 2: eG Enterprise includes comprehensive monitoring for Docker

Learn More: