From e67325cdaf8ce85342dab45b06dbc286c77a5555 Mon Sep 17 00:00:00 2001
From: Yang Yang <yangyang62@baidu.com>
Date: Wed, 14 Mar 2018 00:11:32 +0000
Subject: [PATCH] update readme

---
 doc/design/parallel_executor.md | 42 +++++++++++++++++++++++++--------
 1 file changed, 32 insertions(+), 10 deletions(-)

diff --git a/doc/design/parallel_executor.md b/doc/design/parallel_executor.md
index 567eede1bd5..78ef74f159d 100644
--- a/doc/design/parallel_executor.md
+++ b/doc/design/parallel_executor.md
@@ -30,23 +30,45 @@ operator run on each GPU, it will automatically sync with different streams when
 // if op's input is params' grad:
     // sync with allreduce stream
     // e.g. sgd should wait for allreduce to be finished
-SyncMultipleStreams(op);
+CallBack->BeforeOp(op);
 
 op->Run(*local_scope, place_);
 
 // if op's output is params' grad:
 //     sync with computation stream
 //     e.g. allreduce shoudl wait for fc_grad to be finished.
-SyncMultipleStreams(op);
+CallBack->AfterOp(op);
 ```
 
+And the `Callback` object can be implemented as the following
 
-## API
+```c++
+struct AllReduceCallBack {
+  void BeforeOp(framework::OperatorBase* op);
+  void AfterOp(framework::OperatorBase* op);
+
+  std::unordered_set<std::string> reduced_param_grad_names;
+  std::unordered_set<std::string> param_grad_names_;
+
+  platform::DeviceContext* computation_dev_ctx;    // computation device context
+  platform::DeviceContext* communication_dev_ctx;  // communication device context
 
-The `ParallelExecutor.run` has similar interface as `Executor.run`. Besides
-1. Scope: we don't expose `scope` in `ParallelExecutor.run` since `ParallelExecutor` has its
-own scope to maintain NCCL.
-1. Feed: we don't expose `feed` in the API either, because the whole point of implementing
-parallel_executor is the speed. The input for NN should be implemented in an reader OP.
-1. Fetch: we return the fetched value on all GPUs as a list. (e.g. `exe.run(..., fetch=loss)`
-with return `[loss_on_gpu0, loss_on_gpu1]`)
+  framework::Scope* scope;
+  platform::NCCL::Communicator* nccl_com;
+};
+
+AllReduceCallBack::BeforeOp(framework::OperatorBase* op) {
+  if (op->Input() in reduced_param_grad_names) {
+    communication_dev_ctx->Wait();
+    reduced_param_grad_names.erase(op->Input())
+  }
+}
+
+AllReduceCallBack::AfterOp(framework::OperatorBase* op) {
+  if (op->Output() in param_grad_names) {
+    computation_dev_ctx->Wait();
+    reduced_param_grad_names.insert(op->Output());
+    ncclAllreduce(scope, op->Output(), communication_dev_ctx);
+  }
+}
+```
-- 
GitLab