Repository navigation
/next: drop the streamed time-series results promise; the NDJSON switch measured mixed and stays off - #299
Merged
Conversation
…ch measured mixed and stays off With the merged whole-body reader at ts100 (13 clean interleaved pairs each), BENCH_ARCADEDB_TS_NDJSON=1 helped the served documents arm (0.853 [0.713, 0.928]) and hurt the served native arm (1.087 [1.083, 1.088]); rows identical. The switch stays, default off; the page no longer says the next measurement streams these answers. CAMPAIGN row 85 records the measurement. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Follow-up to #293. That PR added
BENCH_ARCADEDB_TS_NDJSON=1(default off) and a /next sentence plus next-measurement line saying the next measurement reads the served ArcadeDB time-series answers as a stream.The switch has now been measured with the merged reader, which loads the whole de-chunked body before parsing: the lane's own adapter classes from main, the ts100 corpus, the high-usage query, server and client on separate cores, 13 clean interleaved pairs per arm. Documents arm: 0.853 [0.713, 0.928], better. Native arm: 1.087 [1.083, 1.088], worse in all 13 pairs. Identical rows in every run. A mixed result is not something to turn on, so the switch stays off at 26.11.1 and the page must not promise it.
This removes the table sentence and the next-measurement line (and their now unused helpers) from
export_web.py, replaces the two page tests with one that asserts the page says nothing about streaming, and records the measurement in CAMPAIGN section 7 row 85. The switch, its row stamp, the override entry, and the PROTOCOL row stay as they were. Benchmark suite: 648 passed.🤖 Generated with Claude Code