NVIDIA / NVIDIA/cuvs

Tiered index converts to padded on every search call

Open
#2,434 0 comments 0 reactions 1 assignee View on GitHub

@aamijar is already working on this.

Since Aug 12, 2026.

Dominant language
Cuda
Stars
854
Forks
236
Avg merge
3d 3h
Merged PRs (30d)
62

Description

Every tiered index search() call does an internal conversion to padded dataset first.

One solution is to make the owning storage of tiered index padded. New vectors as part of the initial build or extend can still accept standard datasets but can internally be copied into the padded storage.

During search, tiered index currently calls "attach_dataset" (not directly but through a helper called convert_standard_to_padded_index). We are planning to remove these helpers and the "attach_dataset" shortly in favor of update_dataset.

Contributor guide

Open the contributing guide

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.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.