Implement concurrent highlighting on multiple threads (#64) #429
Conversation
This file contains 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
jbaiter
changed the title
jbaiter
force-pushed
the
concurrent-highlighting
branch
from
c2fca29 to
51bd67e
Compare
schmika
reviewed
May 10, 2024
schmika
reviewed
May 10, 2024
jbaiter
force-pushed
the
concurrent-highlighting
branch
from
51bd67e to
cd48938
Compare
- Concurrency is per-(doc, field) combination, i.e. when highlighting 2 field in 10 documents, 20 threads should be used - Thread number and maximum number of queued highlighting tasks can be controlled with the new `numHighlightingThreads` and `maxQueuedPerThread` attributes on the `OcrHighlightComponent` config. - Includes a benchmarking suite that runs against the example index (`example/bench.py`)
jbaiter
force-pushed
the
concurrent-highlighting
branch
from
cd48938 to
527f91d
Compare
jbaiter
force-pushed
the
concurrent-highlighting
branch
from
527f91d to
886d31a
Compare
schmika
reviewed
May 10, 2024
We were storing the number of matches in a non-thredsafe `HashMap`, which led to a race condition when writing to it from multiple threads. We now use a threadsafe `ConcurrentHashMap` to avoid this. Additionally, we now use `CompleteableFuture`s for waiting on our asynchronous highlighting threads instead of raw `Future`s.
Now that we support concurrent highlighting, it's no longer needed.
jbaiter
force-pushed
the
concurrent-highlighting
branch
from
6623265 to
4d1f5a0
Compare
schmika
approved these changes
May 23, 2024
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Outdated suggestions cannot be applied.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.
numHighlightingThreadsandmaxQueuedPerThreadparameters✨ Performance Gains ✨ on a machine with 24 logical cores and enough RAM to hold the index and OCR data in the page cache (i.e. I/O does not play a relevant role):