Enormous histograms

[lgray]

Cluster distributed histograms many 10s to 100s of GB in size with billions of bins are possible:

• proposal: new distributed histogram type dask-contrib/dask-histogram#168

However, at present they are a little slow and could be improved with auxiliary data shuffling services in the filling step.

• What is the actual extent of uses cases for histograms this large?
• How do we make a good user interface to "virtual" histograms that are only ever rendered fully in memory back at the client machine?
• GPUs? GPU (CUDA)-kernel for end-user analysis #49

[lgray]

@ianna I mean - this is an open research question, we haven't exactly solved it.

[ianna]

@ianna I mean - this is an open research question, we haven't exactly solved it.

@lgray - agree. We have closed the issues for the PyHEP 2025 workshop. We ought to (re-)open them for the next one :-)

[eduardo-rodrigues]

Hello. TBH with you I felt the same for at least #55, which is a big and important topic. Having a 2026 label would make me happy, thanks :).