Understanding how B-Tree indexes are built

Hdf-forum archives
1

Hello,

I'm interested in better understanding the internals of how chunks are
indexed inside HDF5 files. I've found general information about chunking
and indexing on the HDF website and other places, but nothing specific on
how the B-Tree indexes are built. In particular I'm interested in knowing
how the index is built to account for the different dimensions in chunks.
If anyone knows about documents or papers covering this it would be very
much appreciated.

Cheers,
Pablo

2

Not sure if you've seen it already but

https://support.hdfgroup.org/HDF5/doc/Advanced/Chunking/index.html

Is a good start.

Mark

"Hdf-forum on behalf of Pablo Rozas Larraondo" wrote:

Hello,

I'm interested in better understanding the internals of how chunks are indexed inside HDF5 files. I've found general information about chunking and indexing on the HDF website and other places, but nothing specific on how the B-Tree indexes are built. In particular I'm interested in knowing how the index is built to account for the different dimensions in chunks. If anyone knows about documents or papers covering this it would be very much appreciated.

Cheers,
Pablo