Skip to content

[CFP ] Containers are just processes: a Linux perspective #197

Description

@ghanatava

Description

Containers are ordinary Linux processes that are started with a different set of kernel primitives primarily namespaces, cgroups, and filesystem isolation. This talk presents a kernel first mental model of containers, focusing on what actually happens when a container starts, rather than how container tools abstract it.

In this 20-minute session, we will walk through how the Linux kernel enables containers by:

  • Creating isolated views of the system using namespaces (PID and mount)
  • Exposing reality through /proc
  • Safely changing the root filesystem using pivot_root
  • pivot_root v/s chroot
  • Limiting resource usage with cgroups

The goal of this talk is not to deep-dive into Docker or Kubernetes internals, but to demystify containers by understanding the exact Linux mechanisms they rely on. By the end of the session, attendees should be able to reason about containers as constrained processes rather than opaque runtime objects.

This talk is aimed at Linux users and engineers who run containers regularly and want a clearer understanding of the kernel primitives underneath no kernel hacking experience required.


Talk Duration

20 minutes


Audience Level

Beginner to Intermediate
(Linux users, DevOps engineers, students)


Prerequisites

Basic familiarity with Linux processes and the command line

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions