Bug #81814: InnoDB adaptive hash index uses a bad partitioning algorithm for the real world #122
Conversation
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
…thm for the real world This patch replaces hash-based AHI partitioning with index ID based one. Index ID based partitioning is better than the current bad hash function. It is also not any worse than any 'good' hash function for real world workloads. Adding space IDs would only make sense if index IDs were not guaranteed to be unique. They are unique in 5.7 and the current public release of 8.0.
|
Hi, thank you for your contribution. Please confirm this code is submitted under the terms of the OCA (Oracle's Contribution Agreement) you have previously signed by cutting and pasting the following text as a comment: |
|
I confirm the code being submitted is offered under the terms of the OCA, and that I am authorized to contribute it. |
|
Hi, thank you for your contribution. Your code has been assigned to an internal queue. Please follow |
laurynas-biveinis
added a commit
to laurynas-biveinis/percona-server
that referenced
this pull request
Apr 3, 2017
…ioning algorithm for the real world) by applying a patch by Alexey Kopytov at mysql/mysql-server#122: Bug #81814: InnoDB adaptive hash index uses a bad partitioning algorithm for the real world This patch replaces hash-based AHI partitioning with index ID based one. Index ID based partitioning is better than the current bad hash function. It is also not any worse than any 'good' hash function for real world workloads. Adding space IDs would only make sense if index IDs were not guaranteed to be unique. They are unique in 5.7 and the current public release of 8.0. (cherry picked from commit d0fb95d1cbab6670b508145954e57a1fe3383885)
gunnarku
pushed a commit
to facebook/mysql-8.0
that referenced
this pull request
Jul 21, 2017
Summary: RocksDB has some static object dependencies, and without following dependencies, it hits SIGABRT. This diff has two fixes. 1. Updating RocksDB revision. D51789 in RocksDB changes the problematic mutex from global variable to local (function-level) static variable. 2. Changing MyRocks no to call ddl_manager.persist_stats() by kill_server_thread (which calls rocksdb_done_func()). By calling ddl_manager.persist_stats(), rocksdb::ThreadLocalPtr::StaticMeta::OnThreadExit() (which calls pthread_mutex_lock on "static port::Mutex mutex") is called at the end of the thread. On the other hand, there is a race between main thread and kill_server_thread. main thread calls pthread_mutex_destroy for the mutex. To guarantee not to call pthread_mutex_lock() after pthread_mutex_destroy(), it is necessary not to call ddl_manager.persist_stats() from kill_server_thread. This diff fixes the issue. @update-submodule: rocksdb Test Plan: mtr cardinality --repeat=100 Reviewers: hermanlee4, spetrunia, jkedgar Reviewed By: jkedgar Subscribers: yhchiang, webscalesql-eng Differential Revision: https://reviews.facebook.net/D51933
gunnarku
pushed a commit
to facebook/mysql-8.0
that referenced
this pull request
Jul 25, 2017
Summary: RocksDB has some static object dependencies, and without following dependencies, it hits SIGABRT. This diff has two fixes. 1. Updating RocksDB revision. D51789 in RocksDB changes the problematic mutex from global variable to local (function-level) static variable. 2. Changing MyRocks no to call ddl_manager.persist_stats() by kill_server_thread (which calls rocksdb_done_func()). By calling ddl_manager.persist_stats(), rocksdb::ThreadLocalPtr::StaticMeta::OnThreadExit() (which calls pthread_mutex_lock on "static port::Mutex mutex") is called at the end of the thread. On the other hand, there is a race between main thread and kill_server_thread. main thread calls pthread_mutex_destroy for the mutex. To guarantee not to call pthread_mutex_lock() after pthread_mutex_destroy(), it is necessary not to call ddl_manager.persist_stats() from kill_server_thread. This diff fixes the issue. @update-submodule: rocksdb Test Plan: mtr cardinality --repeat=100 Reviewers: hermanlee4, spetrunia, jkedgar Reviewed By: jkedgar Subscribers: yhchiang, webscalesql-eng Differential Revision: https://reviews.facebook.net/D51933
dbussink
pushed a commit
to planetscale/mysql-server
that referenced
this pull request
Mar 6, 2024
) (mysql#122) Signed-off-by: Shlomi Noach <[email protected]>
dbussink
pushed a commit
to planetscale/mysql-server
that referenced
this pull request
May 21, 2024
) (mysql#122) Signed-off-by: Shlomi Noach <[email protected]>
dbussink
pushed a commit
to planetscale/mysql-server
that referenced
this pull request
May 21, 2024
) (mysql#122) Signed-off-by: Shlomi Noach <[email protected]>
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.
This patch replaces hash-based AHI partitioning with index ID based
one.
Index ID based partitioning is better than the current bad hash
function. It is also not any worse than any 'good' hash function for
real world workloads.
Adding space IDs would only make sense if index IDs were not guaranteed
to be unique. They are unique in 5.7 and the current public release of
8.0.