[Runtime] [ThreadPool] Make SpscTaskQueue::Pop(..) spin_count configurable #3577
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
|
Thanks @ajtulloch please fix the CI error(was due to compiler warning) |
yidawang
approved these changes
Jul 18, 2019
u99127
reviewed
Jul 19, 2019
ajtulloch
force-pushed
the
threadpool-spin-fix
branch
2 times, most recently
from
9c53c96 to
a5d84b4
Compare
|
ping @ajtulloch |
|
Sorry for the delay, will fix asap and update. |
…rable In cases where we have multiple models or threadpools active, spinning around `sched_yield()` may not be desirable, as it prevents the OS from effectively scheduling other threads. Thus, allow users to conditionally disable this behaviour (via an environment variable `TVM_THREAD_POOL_SPIN_COUNT`, similar to existing environment flags for the thread pool such as `TVM_BIND_THREADS`, etc). This substantially improves tail latencies in some of our multi-tenant workloads in practice. Unit tests have been added - on my laptop, running: ``` TVM_THREAD_POOL_SPIN_COUNT=0 ./build/threading_backend_test; TVM_THREAD_POOL_SPIN_COUNT=1 ./build/threading_backend_test; ./build/threading_backend_test; ``` gives https://gist.github.com/ajtulloch/1805ca6cbaa27f5d442d23f9d0021ce6 (i.e. 97ms -> <1ms after this change)
ajtulloch
force-pushed
the
threadpool-spin-fix
branch
from
a5d84b4 to
eab0cc2
Compare
tqchen
added
status: accepted
and removed
status: need update
|
Thanks @ajtulloch @yidawang @u99127 ! |
wweic
pushed a commit
to wweic/tvm
that referenced
this pull request
Aug 9, 2019
…rable (apache#3577) In cases where we have multiple models or threadpools active, spinning around `sched_yield()` may not be desirable, as it prevents the OS from effectively scheduling other threads. Thus, allow users to conditionally disable this behaviour (via an environment variable `TVM_THREAD_POOL_SPIN_COUNT`, similar to existing environment flags for the thread pool such as `TVM_BIND_THREADS`, etc). This substantially improves tail latencies in some of our multi-tenant workloads in practice. Unit tests have been added - on my laptop, running: ``` TVM_THREAD_POOL_SPIN_COUNT=0 ./build/threading_backend_test; TVM_THREAD_POOL_SPIN_COUNT=1 ./build/threading_backend_test; ./build/threading_backend_test; ``` gives https://gist.github.com/ajtulloch/1805ca6cbaa27f5d442d23f9d0021ce6 (i.e. 97ms -> <1ms after this change)
wweic
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Sep 6, 2019
…rable (apache#3577) In cases where we have multiple models or threadpools active, spinning around `sched_yield()` may not be desirable, as it prevents the OS from effectively scheduling other threads. Thus, allow users to conditionally disable this behaviour (via an environment variable `TVM_THREAD_POOL_SPIN_COUNT`, similar to existing environment flags for the thread pool such as `TVM_BIND_THREADS`, etc). This substantially improves tail latencies in some of our multi-tenant workloads in practice. Unit tests have been added - on my laptop, running: ``` TVM_THREAD_POOL_SPIN_COUNT=0 ./build/threading_backend_test; TVM_THREAD_POOL_SPIN_COUNT=1 ./build/threading_backend_test; ./build/threading_backend_test; ``` gives https://gist.github.com/ajtulloch/1805ca6cbaa27f5d442d23f9d0021ce6 (i.e. 97ms -> <1ms after this change)
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.
In cases where we have multiple models or threadpools active, spinning around
sched_yield()may not be desirable, as it prevents the OS from effectivelyscheduling other threads.
Thus, allow users to conditionally disable this behaviour (via an environment
variable
TVM_THREAD_POOL_SPIN_COUNT, similar to existing environment flags forthe thread pool such as
TVM_BIND_THREADS, etc).This substantially improves tail latencies in some of our multi-tenant
workloads in practice.
Unit tests have been added - on my laptop, running:
gives https://gist.github.com/ajtulloch/1805ca6cbaa27f5d442d23f9d0021ce6 (i.e.
97ms -> <1ms after this change)
Thanks for contributing to TVM! Please refer to guideline https://docs.tvm.ai/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from Reviewers.