Skip to content

Fix memory consumption when max_block_size is huge in Aggregator + more memory checks#51566

Merged
nickitat merged 9 commits intoClickHouse:masterfrom
nickitat:mem_fixes_aggregator
Aug 2, 2023
Merged

Fix memory consumption when max_block_size is huge in Aggregator + more memory checks#51566
nickitat merged 9 commits intoClickHouse:masterfrom
nickitat:mem_fixes_aggregator

Conversation

@nickitat
Copy link
Copy Markdown
Member

@nickitat nickitat commented Jun 28, 2023 •
edited
Loading

Changelog category (leave one):

  • Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Fixed memory consumption in Aggregator when max_block_size is huge.

Fixes: #48173

@robot-ch-test-poll robot-ch-test-poll added the pr-improvement Pull request with some product improvements label Jun 28, 2023
@robot-ch-test-poll
Copy link
Copy Markdown
Contributor

robot-ch-test-poll commented Jun 28, 2023 •
edited by robot-ch-test-poll2
Loading

This is an automated comment for commit a494951 with description of existing statuses. It's updated for the latest CI running
The full report is available here
The overall status of the commit is 🔴 failure

Check nameDescriptionStatus
AST fuzzerRuns randomly generated queries to catch program errors. The build type is optionally given in parenthesis. If it fails, ask a maintainer for help🟢 success
CI runningA meta-check that indicates the running CI. Normally, it's in success or pending state. The failed status indicates some problems with the PR🟡 pending
ClickHouse build checkBuilds ClickHouse in various configurations for use in further steps. You have to fix the builds that fail. Build logs often has enough information to fix the error, but you might have to reproduce the failure locally. The cmake options can be found in the build log, grepping for cmake. Use these options and follow the general build process🟢 success
Compatibility checkChecks that clickhouse binary runs on distributions with old libc versions. If it fails, ask a maintainer for help🟢 success
Docker image for serversThe check to build and optionally push the mentioned image to docker hub🟢 success
Fast testNormally this is the first check that is ran for a PR. It builds ClickHouse and runs most of stateless functional tests, omitting some. If it fails, further checks are not started until it is fixed. Look at the report to see which tests fail, then reproduce the failure locally as described here🟢 success
Flaky testsChecks if new added or modified tests are flaky by running them repeatedly, in parallel, with more randomization. Functional tests are run 100 times with address sanitizer, and additional randomization of thread scheduling. Integrational tests are run up to 10 times. If at least once a new test has failed, or was too long, this check will be red. We don't allow flaky tests, read the doc🟢 success
Install packagesChecks that the built packages are installable in a clear environment🟢 success
Integration testsThe integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests🟢 success
Mergeable CheckChecks if all other necessary checks are successful🟢 success
Performance ComparisonMeasure changes in query performance. The performance test report is described in detail here. In square brackets are the optional part/total tests🔴 failure
Push to DockerhubThe check for building and pushing the CI related docker images to docker hub🟢 success
SQLancerFuzzing tests that detect logical bugs with SQLancer tool🟢 success
SqllogicRun clickhouse on the sqllogic test set against sqlite and checks that all statements are passed🟢 success
Stateful testsRuns stateful functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stateless testsRuns stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stress testRuns stateless functional tests concurrently from several clients to detect concurrency-related errors🟡 pending
Style CheckRuns a set of checks to keep the code style clean. If some of tests failed, see the related log from the report🟢 success
Unit testsRuns the unit tests for different release types🟢 success
Upgrade checkRuns stress tests on server version from last release and then tries to upgrade it to the version from the PR. It checks if the new server can successfully startup without any errors, crashes or sanitizer asserts🟢 success

@nickitat
Copy link
Copy Markdown
Member Author

nickitat commented Jun 29, 2023

aggregation_by_partitions_11_CPU_select a from t_10000000_16 group by a format Null settings optimize_aggregation_in_ord diff

@nickitat nickitat force-pushed the mem_fixes_aggregator branch from e4cfffa to 137263c Compare July 5, 2023 13:15
nickitat
nickitat commented Jul 5, 2023
View reviewed changes
src/Interpreters/Aggregator.h
/// - No two-level aggregation
/// - No external aggregation
/// - No without_key support (it is implemented using executeOnIntervalWithoutKey())
void executeOnBlockSmall(
Copy link
Copy Markdown
Member Author

@nickitat nickitat Jul 5, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure why this methods were in private section since they're used in AggregatingInOrderTransform

nickitat
nickitat commented Jul 5, 2023
View reviewed changes
src/Common/MemoryTracker.cpp
* So, we allow over-allocations.
*/
Int64 will_be = size + amount.fetch_add(size, std::memory_order_relaxed);
Int64 will_be = size ? size + amount.fetch_add(size, std::memory_order_relaxed) : amount.load(std::memory_order_relaxed);
Copy link
Copy Markdown
Member Author

@nickitat nickitat Jul 5, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

rmw atomics are really expensive, this way check() that does alloc(0) will be much cheaper.

@nickitat nickitat changed the title More memory checks + fix memory consumption when max_block_size is huge in Aggregator Fix memory consumption when max_block_size is huge in Aggregator + more memory checks Jul 5, 2023
@nickitat nickitat marked this pull request as ready for review July 5, 2023 13:48
@alexey-milovidov alexey-milovidov self-assigned this Jul 5, 2023
alexey-milovidov
alexey-milovidov reviewed Jul 8, 2023
View reviewed changes
tests/queries/0_stateless/02797_aggregator_huge_mem_usage_bug.sql Outdated
-- this is what allows mem usage to go really high
set max_block_size=10000000000;

set max_memory_usage_for_user = '1Gi';
Copy link
Copy Markdown
Member

@alexey-milovidov alexey-milovidov Jul 8, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why "for user"?
It prevents running the test in parallel.

@alexey-milovidov
Copy link
Copy Markdown
Member

alexey-milovidov commented Jul 8, 2023

@nickitat one integration test looks suspicious.

@nickitat
Copy link
Copy Markdown
Member Author

nickitat commented Jul 24, 2023

@nickitat one integration test looks suspicious.

this test is flaky

@nickitat nickitat merged commit 2cbe79b into ClickHouse:master Aug 2, 2023
@nickitat nickitat mentioned this pull request Aug 7, 2023
Closed
@yl-lisen yl-lisen mentioned this pull request Mar 19, 2024
Closed
25 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@alexey-milovidov alexey-milovidov alexey-milovidov approved these changes

Assignees

@alexey-milovidov alexey-milovidov

Labels

pr-improvement Pull request with some product improvements

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

02151_hash_table_sizes_stats triggers OOM

3 participants

@nickitat @robot-ch-test-poll @alexey-milovidov