According to a study by 451 Research, the application container market will be worth more than $2.1 billion in 2019 and more than $4.3 billion in 2022 — a compound annual growth rate of more than 30 percent. The research firm says this growth is being driven by the need for rapid application development and more efficient infrastructure management to meet digital transformation goals.
Organizations can’t afford to wait months or even years for developers to build monolithic applications that are constrained by infrastructure requirements. Containers enable the delivery of lightweight apps that can be deployed across a variety of operating environments.
Containerization also offers speed and efficiency advantages over virtual machines (VMs). While each VM must run a full copy of the OS, containers allow multiple workloads to run on a single OS instance. Apps simply access the CPU, memory, I/O and network resources of the host OS, bringing significant improvements in startup time and resource utilization.
The accelerated use of containers can cause management issues, however. It is not uncommon for an organization to have scores of loosely coupled containerized components that must work together for a particular app to function as designed. The need to efficiently manage containers across their lifecycle is driving demand for orchestration tools.
An orchestrator determines where a container should run and automatically deploys the container on that host. Typically, an agent is installed on each of the servers or VMs that make up a container platform, and the agent communicates its host’s condition to the orchestrator. In essence, the agent tells the orchestrator, “This host is running and has resources available for containers.” The orchestrator manages this pool of hosts and deploys containers on the best host available. If a host goes down or has inadequate resources, the orchestrator spins up new copies of the containers on other hosts.
Without an orchestrator, your containers would go down when the host goes down. The engine that runs a container on a host doesn’t know about the rest of the environment. The orchestrator understands what hosts are available and moves containers around as needed.
Some orchestrators go beyond this simple “up or down” logic and allow you to establish policies as to how you want your containers to run. For example, you could tell the policy engine, “I want five instances of this container running, and spread them across all of the hosts.” The orchestrator provides the logic and automaton that implements your policy.
Leading orchestration tools such as Kubernetes, Amazon Elastic Container Service, Docker Swarm and Helios offer additional features such as:
- automated deployment and replication of containers
- load balancing of container groups
- management of connections between containers and external services
- configuration of applications in relation to their containers
- rolling upgrades of containers
Kubernetes is the de facto standard in container orchestration. It’s portable and provides a robust feature set that can be delivered to developers in a Platform-as-a-Service paradigm. However, other orchestrators might be a better choice depending on your infrastructure and the specific features you need.
To help with your transition to containerization, contact Technologent. We can perform an assessment of your current environment and help you select the right orchestrator to meet your needs and objectives.
Comments