OpenGrok
Home
Sort by relevance
Sort by last modified time
Full Search
Definition
Symbol
File Path
History
|
|
Help
Searched
refs:RecvAsync
(Results
1 - 12
of
12
) sorted by null
/external/tensorflow/tensorflow/core/common_runtime/
rendezvous_mgr.h
53
void
RecvAsync
(const ParsedKey& key, const Rendezvous::Args& args,
rendezvous_mgr.cc
116
void IntraProcessRendezvous::
RecvAsync
(const ParsedKey& parsed,
122
local_->
RecvAsync
(
rendezvous_util.cc
103
rendezvous->
RecvAsync
(
graph_runner.cc
73
void
RecvAsync
(const ParsedKey& parsed, const Args& recv_args,
/external/tensorflow/tensorflow/core/framework/
rendezvous.h
108
virtual void
RecvAsync
(const ParsedKey& key, const Args& args,
111
// Synchronous wrapper for
RecvAsync
.
rendezvous.cc
121
RecvAsync
(key, recv_args,
196
void
RecvAsync
(const ParsedKey& key, const Args& recv_args,
rendezvous_test.cc
169
// must use
RecvAsync
below. Otherwise, blocking Recv() may run
204
rendez_->
RecvAsync
(MakeKey(strings::StrCat(i)), Rendezvous::Args(),
294
rendez_->
RecvAsync
(
/external/tensorflow/tensorflow/core/kernels/
sendrecv_ops_test.cc
34
void
RecvAsync
(const ParsedKey& key, const Args& args,
sendrecv_ops.cc
183
ctx->rendezvous()->
RecvAsync
(parsed_key_, args,
191
ctx->rendezvous()->
RecvAsync
(in_loop_parsed, args,
/external/tensorflow/tensorflow/core/distributed_runtime/
base_rendezvous_mgr.h
135
void
RecvAsync
(const ParsedKey& key, const Rendezvous::Args& args,
base_rendezvous_mgr.cc
292
void BaseRemoteRendezvous::
RecvAsync
(const ParsedKey& parsed,
296
CHECK(is_initialized()) << "
RecvAsync
called when uninitialized.";
306
local_->
RecvAsync
(
358
local_->
RecvAsync
(parsed, Args(), std::move(done));
/external/tensorflow/tensorflow/contrib/verbs/
README.md
55
The tensor transfer process is initiated when the receiver requests a tensor. In code it is done by calling **Rendezvous::Recv()** or **Rendezvous::
RecvAsync
()**. The TensorFlow base implementation handles the case where the requested tensor is located on the same node. The more interesting case where the requested tensor is located on a remote node (receiver != sender) is to be handled in a derivation of the pure virtual **BaseRemoteRendezvous::RecvFromRemoteAsync()**. TensorFlow provides a default GRPC based implementation which comes in the vanilla version but suffers in scalability when running large models. Our RDMA based implementation presumes to be more scalable. HKUST's contrib GDR implementation is more scalable than GRPC, and less scalable than ours, only because we did our evolution based on it.
Completed in 643 milliseconds