stargz-snapshotter
entr
stargz-snapshotter | entr | |
---|---|---|
10 | 47 | |
1,048 | 4,010 | |
1.7% | - | |
8.4 | 6.8 | |
5 days ago | about 1 month ago | |
Go | C | |
Apache License 2.0 | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
stargz-snapshotter
-
Tree-shaking, the horticulturally misguided algorithm
A lazy chunked delivery strategy like used in the k8s stargz-snapshotter[0] project could be effective here, where it only pulls chunks as needed, but it would probably require wasm platform changes.
[0] https://github.com/containerd/stargz-snapshotter
-
Show HN: depot.ai – easily embed ML / AI models in your Dockerfile
To optimize build speed, cache hits, and registry storage, we're building each image reproducibly and indexing the contents with eStargz[0]. The image is stored on Cloudflare R2, and served via a Cloudflare Worker. Everything is open source[1]!
Compared to alternatives like `git lfs clone` or downloading your model at runtime, embedding it with `COPY` produces layers that are cache-stable, with identical hash digests across rebuilds. This means they can be fully cached, even if your base image or source code changes.
And for Docker builders that enable eStargz, copying single files from the image will download only the requested files. eStargz can be enabled in a variety of image builders[2], and we’ve enabled it by default on Depot[3].
Here’s an announcement post with more details: https://depot.dev/blog/depot-ai.
We’d love to hear any feedback you may have!
[0] https://github.com/containerd/stargz-snapshotter/blob/main/docs/estargz.md
[1] https://github.com/depot/depot.ai
[2] https://github.com/containerd/stargz-snapshotter/blob/main/docs/integration.md#image-builders
[3] https://depot.dev
-
A Hidden Gem: Two Ways to Improve AWS Fargate Container Launch Times
Seekable OCI (SOCI) is a technology open-sourced by AWS that enables containers to launch faster by lazily loading the container image. It’s usually not possible to fetch individual files from gzipped tar files. With SOCI, AWS borrowed some of the design principles from stargz-snapshotter, but took a different approach. A SOCI index is generated separately from the container image and is stored in the registry as an OCI Artifact and linked back to the container image by OCI Reference Types. This means that the container images do not need to be converted, image digests do not change, and image signatures remain valid.
- containerd/stargz-snapshotter: Fast container image distribution plugin with lazy pulling
- EStargz: Lazy pull container images for faster cold starts
- How to optimize the security, size and build speed of Docker images
-
Speeding up LXC container pull by up to 3x
This is interesting and seems general purpose. Not merely for container images.
There’s this option for OCI containers which I don’t pretend to understand: https://github.com/containerd/stargz-snapshotter
It is used by containerd and nerdctl. You do have to build the image with it. Images work in OCI compatible registry. By fetching most used files first container can be started before loading is finished. Or so I gather.
-
Optimizing Docker image size and why it matters
stargz is a gamechanger for startup time. You might not need to care about image size at all
kubernetes and podmand support it, and docker support is likely coming. It lazy loads the filesystem on start-up, making network requests for things that are needed and therefore can often start up large images very fast.
https://github.com/containerd/stargz-snapshotter
-
FOSS News International #2: November 8-145, 2021
containerd/stargz-snapshotter: Fast container image distribution plugin with lazy pulling (github.com)
-
Introducing GKE image streaming for fast application startup and autoscaling
Yes, see https://github.com/containerd/stargz-snapshotter
entr
- Entr – tool for watching files and running commands
-
Meet entr, the standalone file watcher
entr ("Event Notify Test Runner"; GitHub), is a command-line tool written by Eric Radman that allows running arbitrary commands whenever files change.
-
How to build a website without frameworks and tons of libraries
I use something very similar on https://lunar.fyi and https://lowtechguys.com but I wouldn’t call this “simple” anymore.
They use Jinja templating, I prefer Slim (https://github.com/slim-template/slim#syntax-example) which has a more Pythonic syntax (there is plim [0] in Python for that)
I use Tailwind as well for terse styling and fast experimentation (allows me to write a darkMode-aware and responsive 100 line CSS in a single line with about 10 classes)
For interaction I can write CoffeeScript directly in the page [1] and have it compiled by plim.
I run a Caddy static server [2] and use Syncthing [3] to have every file save deployed instantly to my Hetzner server.
I use entr [4] and livereloadx [5] to rebuild the pages and do hot reload on file save. All the commands are managed in a simple Makefile [6]
———
You can already see how the footnotes take up a large chunk of this comment, this is not my idea of simple. Sure, the end result is readable static HTML and I never have to fight obscure React errors, but it’s a high effort setup for starters.
Simple for me would be: write markdown files for pages, a simple CSS for general styling (should be optional), click to deploy on my domain. Images should automatically be resized to multiple sizes and optimized, videos re-encoded for smaller filesize etc.
I have mostly implemented that for myself (https://notes.alinpanaitiu.com/How%20I%20write%20this%20blog...) but it feels fragile. I’d rather pay for a professional solution.
[0] https://plim.readthedocs.io/en/latest/
[1] https://github.com/FuzzyIdeas/lowtechguys/blob/main/src/rcmd...
[2] https://caddyserver.com/docs/command-line#caddy-file-server
[3] https://syncthing.net
[4] https://github.com/eradman/entr
[5] https://nitoyon.github.io/livereloadx/
[6] https://github.com/FuzzyIdeas/lowtechguys/blob/main/Makefile
- How to start a Go project in 2023
-
[Guide] A Tour Through the Python Framework Galaxy: Discovering the Stars
Try entr for fast reloading. Another one is hupper.
- Use entr when working on you rice for auto config refreshing
- The Unix process API is unreliable and unsafe
- How do you develop cloud-native applications locally on Kubernetes?
-
What are the not-so-obvious tools that you don't want to miss?
entr
- Test driven development is adhd dream
What are some alternatives?
kube-fledged - A kubernetes operator for creating and managing a cache of container images directly on the cluster worker nodes, so application pods start almost instantly
watchexec - Executes commands in response to file modifications
acr - Azure Container Registry samples, troubleshooting tips and references
nextjs-tailwind-ionic-capacitor-starter - A starting point for building an iOS, Android, and Progressive Web App with Tailwind CSS, React w/ Next.js, Ionic Framework, and Capacitor
containerd - An open and reliable container runtime
modd - A flexible developer tool that runs processes and responds to filesystem changes
soci-snapshotter - A containerd snapshotter plugin which enables standard OCI images to be lazily loaded without requiring a build-time conversion step.
swc-node - Faster ts-node without typecheck
snoop - Snoop — инструмент разведки на основе открытых данных (OSINT world)
air - ☁️ Live reload for Go apps
uChmViewer - A fork of Kchmviewer, the best software for viewing .chm (MS HTML help) and .epub eBooks.
vim-test - Run your tests at the speed of thought