-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
49 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,49 @@ | ||
| --- | ||
| title: "Silly Project Ideas" | ||
| date: 2025-01-10T21:28:12-08:00 | ||
| tags: | ||
| - dev | ||
| - go | ||
| - beam | ||
| - prism | ||
| - hobby SDK | ||
| categories: | ||
| - Dev | ||
| - Hobby SDK | ||
| --- | ||
|
|
||
| Here are some weird project Ideas I'd love to try if I ever find the time. | ||
|
|
||
| <!--more--> | ||
|
|
||
| * Make [Prism](https://beam.apache.org/documentation/runners/prism/) "self distributed and scaling" using [Cloud Run](https://cloud.google.com/run). | ||
| * Basically the idea would be to host an instance on Cloud Run, serving it's JobManagement service. | ||
| * But replace the local docker execution with Cloud Run launches for worker container launches, connecting back to the parent prism instance for work orchestration. | ||
| * The main prism instance can't scale, but the UI should be hostable without too much problem. | ||
| * Could add Tailscale for more fun. | ||
| * Entirely a fun hack without also adding a cloud persistence layer too. | ||
| * Make [Prism] integrate with Kubernetes. Distributes work in the cluster. | ||
| * Basically the same idea, but with Kubernetes operations instead. | ||
| * Same problems. | ||
| * Find the branch of the Go compiler that has the work for Memory/GC Regions, and have it store things into Cuda Unified Memory Instead to simplify GPU integration. | ||
| * Totally wild endgame idea. Would need to actaully learn some Cuda and it's CGo bindings first. | ||
| * In principle could be used to more naturally toggle GPU based work. | ||
| * Heck, even just do my normal distributed ray tracer, but batch the rendering stage onto a GPU. | ||
| * Try using opencv with BeamGo generally. | ||
| * Actually finish the Hobby SDK. | ||
| * Stateful Transforms - State and Timers | ||
| * On Window Expiration | ||
| * Real Windowing | ||
| * Streaming Splittable-DoFns | ||
| * Leverage Go 1.24 Generic Aliases to make the API more convenient. | ||
| * And of course | ||
| * Submit talk proposals to conferences. | ||
|
|
||
| So might to potentially do. | ||
|
|
||
| Never enough time. | ||
|
|
||
| ---- | ||
|
|
||
| I think I'm not going to try too hard to post on Thursdays. | ||
| Between the Frelard Run Club, and Pub Night, it's quite a packed day. |