Skip to content

Docs: the JVM Arrow vector load is faster and uses more memory than the 26.11.1 route (refs #348) - #367

Merged
tae898 merged 1 commit into
mainfrom
docs/jvm-arrow-vector-memory
Nov 10, 2026
Merged

tae898 merged 1 commit into
mainfrom
docs/jvm-arrow-vector-memory

Conversation

@tae898

@tae898 tae898 commented Nov 10, 2026

Copy link
Copy Markdown
Collaborator

Refs #348. I added one bullet to the section "Vertex types in insert_dataframe and insert_arrow" of the bulk loads page. On arcadedb-jvm, a float32 vector column of such a load crosses to Java as one matrix for each batch. The A/B of that route against the one of 26.11.1 finished on 2026-11-11: it is faster in all six gating figures, and the peak memory of the load is higher than on the old route, so the page now says both and points to #348, where the measurement is. I did not add figures or advice about batch_size, because the study does not measure either.

The docs build (the command of "Build docs site") exits 0 with no warnings, and the added line has no em dashes and no pip install.

🤖 Generated with Claude Code

…ry than the 26.11.1 route (refs #348)

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@tae898
tae898 merged commit 2badf0f into main Nov 10, 2026
5 checks passed
@tae898
tae898 deleted the docs/jvm-arrow-vector-memory branch November 10, 2026 19:42
@github-actions github-actions Bot added this to the 26.12.1 milestone Nov 10, 2026
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

Successfully merging this pull request may close these issues.

1 participant