Ryan Abernathey

150 GB / 295 s = 508 MB/s That's not a bad throughput and might getting close to the I/O limit. I assume you've got an SSD. Some SSDs can...

Hi @fecampos thanks for sharing. Could you please share the full code and data you used to produce this figure? Without that it's impossible for us to help you.

Hi @yarikoptic, thanks for sharing this! Looks cool! FYI, we are planning on open sourcing the solution we have built at Earthmover later this fall.

Hi folks! We released our project! You can read all about it here: https://icechunk.io/

Later this spring, Earthmover will be launching a free tier which will provide a catalog service for public Icechunk datasets. (Similar to GitHub's free tier for public git repos.) We...

Thanks for sharing this Tom. It has been great to have you spending time on Zarr recently and bringing a fresh perspective to long-standing discussions. FWIW, I'm on record in...

> The simplest way to think of this is that the store is kind of like a file object, opening it in `w` mode will give you a new (empty)...

This attribute needs to just be coerced to a plain int.

I think we should make the defaults exactly the same as they are in v2. I do not think we should make sharding the default until we have spent more...

> @rabernat can you expand a bit on why you think we should keep the same defaults? Just for the sake of maintaining consistency and not having to make a...