-
Notifications
You must be signed in to change notification settings - Fork 25
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Investigate using Crossbeam lock-free queue + threads #30
Comments
56quarters
changed the title
Investigate using Crossbean lock-free queue + threads
Investigate using Crossbeam lock-free queue + threads
Dec 7, 2016
Performance looks promising so far.
The "cb" benchmark is the one that uses crossbeam. |
56quarters
added a commit
that referenced
this issue
Dec 19, 2016
Implement an async MetricSink (QueuingMetricSink) that sends metrics to a wrapped sink that runs in a separate thread. Communicate between the sink and separate thread via a lock-free Crossbeam queue. Fixes #30
56quarters
added a commit
that referenced
this issue
Dec 19, 2016
Implement an async MetricSink (QueuingMetricSink) that sends metrics to a wrapped sink that runs in a separate thread. Communicate between the sink and separate thread via a lock-free Crossbeam queue. Fixes #30
56quarters
added a commit
that referenced
this issue
Dec 19, 2016
Implement an async MetricSink (QueuingMetricSink) that sends metrics to a wrapped sink that runs in a separate thread. Communicate between the sink and separate thread via a lock-free Crossbeam queue. Fixes #30
56quarters
added a commit
that referenced
this issue
Dec 21, 2016
Implement an async MetricSink (QueuingMetricSink) that sends metrics to a wrapped sink that runs in a separate thread. Communicate between the sink and separate thread via a lock-free Crossbeam queue. Fixes #30
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
There seems to be a fair amount of locking going on in the threadpool we're using. This, combined with how short the actual tasks to run are mean using a thread pool doesn't actually seem to make anything faster.
Let's investigate using the lock-free queue in Crossbeam along with some sort of custom thread worker. It's probably sufficient to only run a single thread worker as well.
http://aturon.github.io/crossbeam-doc/crossbeam/sync/struct.MsQueue.html
The text was updated successfully, but these errors were encountered: