Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Deserializing const_array #228

Open
systemed opened this issue Feb 25, 2024 · 2 comments
Open

Deserializing const_array #228

systemed opened this issue Feb 25, 2024 · 2 comments

Comments

@systemed
Copy link

I'm (belatedly!) moving from zfp 0.5.x to 1.0.1 and am very impressed with the read-only variable rate arrays.

With the fixed-rate arrays it was possible to serialize to disk once and then read them in later, thereby saving time compressing them on application startup.

Is there any way to do this with const_array? The documentation only mentions "zfp’s read-write compressed arrays can be serialized to sequential, contiguous storage".

I'm not too worried about the headers (I can store that information independently) but being able to populate compressed_data from disk would be really useful.

@lindstro
Copy link
Member

I'm afraid we didn't have time to work out (de)serialization with the const_array classes for the 1.0.x releases, but it is likely that this will be supported in the next release. We have to deal with the same issue to support parallel variable-rate decompression, where we need to quickly locate any given block in the compressed stream. We're essentially using the same block offset coding strategy for that. Because we cannot embed these block offsets in the compressed stream while retaining backwards compatibility, we need for the block index to be stored separately, and we're currently working on support for (de)serializing this auxiliary data.

@systemed
Copy link
Author

That's great to know - thank you. I'll go with compressing at application startup for now, but move to deserialization when that's available.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants