Remove LinkedDeque and replace with LinkedHashMap#6968
Merged
andrross merged 2 commits intoopensearch-project:mainfrom Apr 5, 2023
Merged
Remove LinkedDeque and replace with LinkedHashMap#6968andrross merged 2 commits intoopensearch-project:mainfrom
andrross merged 2 commits intoopensearch-project:mainfrom
Conversation
After the recent changes the usage of the LinkedDeque fits quite well with the insertion order semantics of LinkedHashMap, which also allows for constant time additions and removals. Signed-off-by: Andrew Ross <andrross@amazon.com>
andrross
requested review from
Bukhtawar,
CEHENKLE,
Rishikesh1159,
VachaShah,
anasalkouz,
dblock,
dreamer-89,
gbbafna,
kartg,
kotwanikunal,
mch2,
nknize,
owaiskazi19,
reta,
ryanbogan,
saratvemulapalli,
setiah,
shwetathareja,
tlfeng and
xuezhou25
as code owners
Contributor
Gradle Check (Jenkins) Run Completed with:
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #6968 +/- ##
============================================
- Coverage 70.66% 70.57% -0.10%
+ Complexity 59231 59071 -160
============================================
Files 4812 4812
Lines 283761 283617 -144
Branches 40917 40909 -8
============================================
- Hits 200519 200153 -366
- Misses 66784 67033 +249
+ Partials 16458 16431 -27 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Contributor
|
@andrross I like it, the implementation is much simpler now, interestingly enough, the |
reta
reviewed
Apr 4, 2023
| private final StatsCounter<K> statsCounter; | ||
|
|
||
| private volatile ReentrantLock lock; | ||
| private final ReentrantLock lock; |
Contributor
There was a problem hiding this comment.
👍 we could replace
final ReentrantLock lock = this.lock;
lock.lock();
with just
lock.lock();
now
Member
Author
There was a problem hiding this comment.
Oh good catch! I'm sure there was some previous incarnation of this code where a volatile lock was needed, but it definitely was never necessary here.
reta
approved these changes
Apr 4, 2023
Signed-off-by: Andrew Ross <andrross@amazon.com>
Contributor
Gradle Check (Jenkins) Run Completed with:
|
Member
Author
|
@reta regarding |
Contributor
Gradle Check (Jenkins) Run Completed with:
|
Contributor
Gradle Check (Jenkins) Run Completed with:
|
mitrofmep
pushed a commit
to mitrofmep/OpenSearch
that referenced
this pull request
Apr 5, 2023
…#6968) * Remove LinkedDeque and replace with LinkedHashMap After the recent changes the usage of the LinkedDeque fits quite well with the insertion order semantics of LinkedHashMap, which also allows for constant time additions and removals. Signed-off-by: Andrew Ross <andrross@amazon.com> * Use class member reference now that lock is final Signed-off-by: Andrew Ross <andrross@amazon.com> --------- Signed-off-by: Andrew Ross <andrross@amazon.com> Signed-off-by: Valentin Mitrofanov <mitrofmep@gmail.com>
Member
kotwanikunal
pushed a commit
to kotwanikunal/OpenSearch
that referenced
this pull request
May 17, 2023
…#6968) * Remove LinkedDeque and replace with LinkedHashMap After the recent changes the usage of the LinkedDeque fits quite well with the insertion order semantics of LinkedHashMap, which also allows for constant time additions and removals. Signed-off-by: Andrew Ross <andrross@amazon.com> * Use class member reference now that lock is final Signed-off-by: Andrew Ross <andrross@amazon.com> --------- Signed-off-by: Andrew Ross <andrross@amazon.com> (cherry picked from commit 65443ad)
6 tasks
reta
pushed a commit
that referenced
this pull request
May 17, 2023
… filecache support in clear indices cache API (#7595) * Remove LinkedDeque and replace with LinkedHashMap (#6968) * Remove LinkedDeque and replace with LinkedHashMap After the recent changes the usage of the LinkedDeque fits quite well with the insertion order semantics of LinkedHashMap, which also allows for constant time additions and removals. Signed-off-by: Andrew Ross <andrross@amazon.com> * Use class member reference now that lock is final Signed-off-by: Andrew Ross <andrross@amazon.com> --------- Signed-off-by: Andrew Ross <andrross@amazon.com> (cherry picked from commit 65443ad) * Add filecache support in clear indices cache API (#7498) Signed-off-by: Kunal Kotwani <kkotwani@amazon.com> (cherry picked from commit a1e42b1) Signed-off-by: Kunal Kotwani <kkotwani@amazon.com> --------- Signed-off-by: Kunal Kotwani <kkotwani@amazon.com> Co-authored-by: Andrew Ross <andrross@amazon.com>
Member
|
Looks like #7595 includes this? @kotwanikunal |
Member
|
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
5 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.
After the recent changes the usage of the LinkedDeque fits quite well with the insertion order semantics of LinkedHashMap, which also allows for constant time additions and removals.
Microbenchmark Results
The benchmark results are quite similar. The LinkedHashMap version seems to be slightly faster, but they are close enough there could be other variables in the test runs causing the difference. The benchmarks show there isn't any significant regression.
OS: Ubuntu 20.04.5 LTS
Host: AWS EC2 c6i.8xlarge
Benchmark command:
./gradlew -p benchmarks run --args 'FileCacheBenchmark'This PR:
Previous implementation:
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.