The latest articles on DEV Community by myzn (@hellomyzn). https://dev.to/hellomyzn https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F228208%2Fa44ee614-a9c4-4365-8a1a-17b1a37e73f4.JPG https://dev.to/hellomyzn en myzn Sat, 12 Apr 2025 03:47:06 +0000 https://dev.to/hellomyzn/docker-vs-podman-why-the-future-is-rootless-and-daemonless-5488 https://dev.to/hellomyzn/docker-vs-podman-why-the-future-is-rootless-and-daemonless-5488 <h2> Key Takeaways </h2> <ul> <li>Docker relies on a root-privileged daemon (<code>dockerd</code>), while Podman operates daemonless.</li> <li>Podman supports full Rootless container management out of the box.</li> <li>Docker requires additional setup to integrate with Kubernetes, while Podman is natively CRI-O compatible.</li> <li>With Podman, containers are managed as independent Linux processes — no always-on background service needed.</li> <li>Rootless Podman significantly reduces the security risks associated with traditional Docker setups.</li> <li>If <code>dockerd</code> crashes, Docker containers may be impacted; Podman containers remain unaffected.</li> <li>Modern container orchestration trends (like Kubernetes) increasingly favor daemonless runtimes like Podman.</li> </ul> <p>👉 <strong><a href="https://medium.com/@hellomyzn/docker-vs-podman-understanding-the-shift-toward-rootless-daemonless-containers-c0bd9a58244f" rel="noopener noreferrer">Read the full article on Medium</a></strong></p> <p>Stay tuned for more deep dives into the evolving world of container technology! 🚀</p> docker podcast myzn Sat, 12 Apr 2025 03:35:19 +0000 https://dev.to/hellomyzn/welcome-to-the-era-of-1-llm-per-dev-environment-with-docker-model-runner-58l2 https://dev.to/hellomyzn/welcome-to-the-era-of-1-llm-per-dev-environment-with-docker-model-runner-58l2 <h2> Quick Summary </h2> <ul> <li>Docker Model Runner lets you spin up LLMs like Docker containers.</li> <li>One command: <code>docker model run &lt;model-name&gt;</code> does it all.</li> <li>No need for manual server code or dependency setup.</li> <li>Automatically exposes OpenAI-compatible endpoints.</li> <li>REST API makes it easy to connect with any web server (Flask, FastAPI, Next.js, etc.).</li> <li>Simple integration with <code>docker-compose</code>.</li> <li>Currently in closed beta (CPU only, GPU support coming soon).</li> </ul> <p>For a detailed breakdown of how to integrate Docker Model Runner with your own web application, check out the full article here:</p> <p>👉 <a href="https://medium.com/@hellomyzn/welcome-to-the-era-of-1-llm-per-dev-environment-with-docker-model-runner-6a7ac6946379" rel="noopener noreferrer">Read the full article on Medium</a></p> <p>Stay tuned — it's an exciting time for AI infrastructure!</p> docker llm