hyper-parquet: Fix caching of Parquet files and add single-file variant#969
Closed
caetanosauer wants to merge 4 commits into
Closed
hyper-parquet: Fix caching of Parquet files and add single-file variant#969caetanosauer wants to merge 4 commits into
caetanosauer wants to merge 4 commits into
Conversation
Fixes ClickHouse#936. The shared driver (lib/benchmark-common.sh) calls ./query once per try and, for daemon-backed systems, keeps the server alive across tries so tries 2..N measure hot execution. Hyper's ./query instead opened a brand-new HyperProcess on every call, so each "hot" try hit an empty buffer pool against a just-cache-dropped file: every reported hot time was actually cold. Convert both hyper/ and hyper-parquet/ to the client-server model the framework expects (mirroring umbra/): - start: background a supervisor that opens one long-lived hyperd and publishes its connection descriptor to server.endpoint. In hyper/ it also holds a keep-alive connection to hits.hyper so the buffer pool isn't torn down when each per-try ./query process exits (Hyper detaches a .hyper DB when its last connection closes). - stop: SIGTERM the supervisor (cleanly shutting down hyperd) and wait for it to fully exit so drop_caches isn't defeated by pinned mmap pages. - check / query / load: reconnect to the persistent server via its descriptor instead of spawning their own HyperProcess. Loading through the same server also avoids briefly running two hyperd instances (each claiming ~80% RAM) during the heavy COPY. - benchmark.sh: BENCH_RESTARTABLE=yes (there is now a real daemon whose lifecycle matters) and drop the BENCH_CONCURRENT_DURATION=0 override, re-enabling the concurrent-QPS test. Net effect: the driver's cold cycle (stop -> wait -> drop_caches -> start) gives an honest cold try 1, and tries 2..N hit the warm server = genuinely hot. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
caetanosauer
changed the title
The hyper-parquet harness created a TEMP external table per ./query connection. With the persistent-server model (issue ClickHouse#936) the driver opens a fresh connection per try. This caused our plan cache to be wiped, requiring re-sampling the data on every hot run as well. To properly test a cached scenario, this commit changes our harness to create persistent external tables and maintain an attached database throughout the server's lifetime. The query iterations themselves still open a fresh connection each time. We also introduce the `single` variant for the parquet lane, now reporting both `hyper-parquet-single` and `hyper-parquet-partitioned`. As a drive-by change, we also drop ::bigint casts from ClickBench Q29, since those are not needed anymore, matching the queries used by other systems as well. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
caetanosauer
force-pushed
the
hyper-parquet-persistent-external-plancache
branch
from
341179e to
1bc7035
Compare
Contributor
Author
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
2 participants
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.
The hyper-parquet harness created a TEMP external table per ./query
connection. With the persistent-server model (issue #936) the driver
opens a fresh connection per try. This caused our plan cache to be wiped,
requiring re-sampling the data on every hot run as well. To properly
test a cached scenario, this commit changes our harness to create
persistent external tables and maintain an attached database throughout
the server's lifetime. The query iterations themselves still open a
fresh connection each time.
We also introduce the
singlevariant for the parquet lane, nowreporting both
hyper-parquet-singleandhyper-parquet-partitioned.As a drive-by change, we also drop ::bigint casts from ClickBench Q29,
since those are not needed anymore, matching the queries used by other
systems as well.
Co-Authored-By: Claude Opus 4.8 (1M context) noreply@anthropic.com