#Apache #OpenDAL est une couche d'abstraction au stockage des données, en proposant une interface commune (https://github.com/apache/OpenDAL).
On retrouve le file system, AWS S3, mais aussi Google Drive.

Utile lorsque l'on souhaite supporter plusieurs stockages pour une même fonctionnalité. Et bien entendu avec la bibliothèque #Python qui va avec : https://github.com/apache/opendal/blob/main/bindings/python/README.md

I've found an interesting project: Apache OpenDAL
The premise is that it is a single API and wrapper for all data access (filesystems, S3, WebDAV, FUSE, Key/Value stores like Redis, etc). With language bindings for #java, #python, #nodejs, #rust

https://opendal.apache.org/

#apache #opendal

Welcome to Apache OpenDAL™ | Apache OpenDAL™

OpenDAL represents Open Data Access Layer. Our vision is One Layer, All Storage.

New Blog Post Published.

How to Condense 800 Integration Tests into 80, Losslessly?

In this post, I'll share how I condense the extensive integration test matrix in #OpenDAL from `O(T * S * C)` into a linear `O(T + S + C)` without losing any coverage.

https://xuanwo.io/2024/03-how-to-condense-800-integration-tests-into-80-losslessly/

How to Condense 800 Integration Tests into 80, Losslessly?

An infrastructure engineer, focused on distributed storage system

New Blog Post Published.

Why SQL hang for exactly 940s? TCP and Async Rust!

I debugged an issue that Databend SQL could hang for 940s during #OpenDAL reads. I gained a lot of insight into TCP and Async Rust in the process. Feel free to take a look!

https://xuanwo.io/2024/01-why-sql-hang-for-exactly-940s/

Why SQL hang for exactly 940s? TCP and Async Rust!

Achieving Data Freedom Through Open Source and Rust

We've adjusted the schedule of #OpenDAL Community Sync to include two sessions:

- one at 9:00 AM UTC+8 on Friday
- another at 7:00 PM UTC+8 on Friday.

I expect this will make it more convenient for contributors to attend at least every six weeks.

@zhenboli The GHA Cache logic is implemented in #OpenDAL, but it stores each cache separately to better comply with GitHub's rate limit. I'm not sure if it suitable for big cache.

@xuanwo

I have a question about your work #sccache and #OpenDAL.

To determine if the file is cached in GitHub Action[1], the logic is implemented in sccache or opendal?

My X Problem[2]: I want to cache #Nix within #GitHubAction. AFAIK, current solutions (https://github.com/nix-community/cache-nix-action) will consider `/nix` as a whole object, causing "Caches are isolated between branches". Besides, the cost of downloading a big cache is not negligible.

[1]: https://xuanwo.io/reports/2023-04/sccache.png
[2]: https://en.wikipedia.org/wiki/XY_problem

GitHub - nix-community/cache-nix-action: Cache Nix Store in GitHub Actions to speed up workflows [maintainer=@deemp]

Cache Nix Store in GitHub Actions to speed up workflows [maintainer=@deemp] - nix-community/cache-nix-action

GitHub

#OpenDAL Community Sync (testing) is scheduled for 2:00 PM on September 1st, UTC+8

I'm very excited about this meeting and I hope to see you there!

More details: https://lists.apache.org/thread/1vgoq71nw73qxhh05j7tsb2dg36v0d80

I feel incredibly happy when my project #OpenDAL been acknowledged.
I'm building somthing interesting like #OpenDAL + #postgresql