llnl / llnl/conduit

relay: investigate h5literate vs simple walk performance

Open
#391 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

performance
Dominant language
C++
Stars
248
Forks
72
Avg merge
1d 16h
Merged PRs (30d)
15

Description

we use h5literate in relay's hdf5 read functions to follow a pattern that breaks any cycles in an HDF5 graph when reading in data.

Our datasets are usually trees (especially those written by conduit), so this may be overkill and we may be giving up performance. Analysis is needed to see what could be gained with a simpler traversal strategy.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue points to relay's HDF5 read functions and their use of h5literate. Start by locating those entry points and measuring the current traversal on representative tree-shaped datasets, then compare it with a simple walk while checking cycle behavior. Done means a documented performance comparison and a clear recommendation for whether the traversal should change.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
data-engineering, performance
Issue type
Refactor
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.