aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/core/protobuf
diff options
context:
space:
mode:
authorGravatar A. Unique TensorFlower <gardener@tensorflow.org>2018-09-28 14:08:25 -0700
committerGravatar TensorFlower Gardener <gardener@tensorflow.org>2018-09-28 14:34:43 -0700
commitf83da5b0aa37ba55c1b2eaa093e6d043b73f5982 (patch)
treed28c727251f910dc0c7b7a6184286919d436e88f /tensorflow/core/protobuf
parent1724d155f00b49bc817189247cbfb0df2092a9da (diff)
Introduce the abstraction of RunHandler which each DirectSession can use for
the duration of a single RunInternal() call from RunHandlerPool. It is used for running inter-op closures with a global scheduler (which in the future) to improve both median and tail latency (for use-cases like CPU inference). In the case that global pools aren't used, this change should be a no-op. PiperOrigin-RevId: 214992852
Diffstat (limited to 'tensorflow/core/protobuf')
-rw-r--r--tensorflow/core/protobuf/config.proto5
1 files changed, 5 insertions, 0 deletions
diff --git a/tensorflow/core/protobuf/config.proto b/tensorflow/core/protobuf/config.proto
index 85cd02350a..104ab039cb 100644
--- a/tensorflow/core/protobuf/config.proto
+++ b/tensorflow/core/protobuf/config.proto
@@ -453,6 +453,11 @@ message RunOptions {
// same group_key value (in a distributed computation where tasks
// run disjoint graphs).
int64 collective_graph_key = 1;
+ // If true, then operations (using the inter-op pool) across all
+ // session::run() calls will be centrally scheduled, optimizing for (median
+ // and tail) latency.
+ // Consider using this option for CPU-bound workloads like inference.
+ bool use_run_handler_pool = 2;
};
Experimental experimental = 8;