follow comments

Author: typhoonzero

doc/design/refactor/distributed_architecture.md renamed to doc/design/dist_refactor/distributed_architecture.md

@@ -52,8 +52,9 @@ The IR for PaddlePaddle after refactoring is called a `Block`, it specifies the
 
 The user can not directly specify the parameter update rule for the parameter server in the Python module, since the parameter server does not use the same computation definition as the trainer. Instead, the update rule is baked inside the parameter server. The user can not specify the update rule explicitly.
 
-This could be fixed by making the parameter server run the same computation definition as the trainer (the user's Python module). For a detailed explanation, refer to this document -
-[Design Doc: Operation Graph Based Parameter Server](./parameter_server.md)
+This could be fixed by making the parameter server also run an IR, which can be different to the trainer side
+For a detailed explanation, refer to this document -
+[Design Doc: Parameter Server](./parameter_server.md)
 
 ## Distributed Training Architecture
 
@@ -113,7 +114,7 @@ Below are the steps that are followed :
    distributed training program:
    1. Parse configurations from `RemoteExecutor`.
    1. Determine the type of distributed program, can be DataParallelism, ModelParallelism or Streaming.
-   1. Partition the `ProgramDesc` according to type and add `send` / `recv` OP pair on the boundaries. For
+   1. Partition the `ProgramDesc` according to type and add `send` / `recv` OP pair on the boundaries. Take
       DataParallelism type for example, it removes the optimization operators and add a `send` OP to the
       "trainer" role, then add the optimization operators to the parameter server role within the `recv` OP.
 1. Dispatch the partitioned graph to different `RemoteExecutor` in the cluster.
@@ -129,12 +130,9 @@ log printing.
 The Python `RemoteExecutor` is derived from `Executor` class.
 
 ```python
-run(self,
-    program=None,
-    feed=None,
-    fetch_list=None,
-    feed_var_name='feed',
-    fetch_var_name='fetch',
+exe = RemoteExecutor(
+    feed=feeder.feed(data),
+    fetch_list=[avg_cost],
     job_desc=JobDesc(
       jobname,
       num_trainer,
@@ -145,6 +143,10 @@ run(self,
       cpu_per_pserver,
       mem_per_pserver
     ))
+for data in train_reader():
+    loss, acc = exe.run(trainer_prog,
+                        feed=feeder.feed(data),
+                        fetch_list=[avg_cost])
 ```
 
 `JobDesc` object describe the distributed job resource specification to run on

doc/design/refactor/multi_cpu.md renamed to doc/design/dist_refactor/multi_cpu.md

@@ -1,4 +1,4 @@
-# Design Doc: Operation Graph Based Parameter Server
+# Design Doc: Parameter Server
 
 ## Abstract
 
@@ -10,7 +10,7 @@ different purposes.
 ## Background
 
 The previous implementations of the parameter server does not run a
-subgraph. parameter initialization, optimizer computation, network
+fluid sub-program. Parameter initialization, optimizer computation, network
 communication and checkpointing are implemented twice on both the
 trainer and the parameter server.
 
@@ -23,18 +23,17 @@ server becomes a natural extension.
 
 ## Design
 
-### Graph Converter
+### Distributed Transpiler
 
-The *graph converter* converts the user-defined operation (OP) graph
-into subgraphs to be scheduled on different nodes with the following
+The *Distributed Transpiler* converts the user-defined fluid program
+into sub-programs to be scheduled on different nodes with the following
 steps:
 
 1. OP placement: the OPs will be placed on different nodes according
    to heuristic that minimizes estimated total computation
    time. Currently we will use a simple heuristic that puts parameter
    varable on parameter server workers and everything else on trainer
    workers.
-
 1. Add communication OPs to enable the communication between nodes.
 
 We will need these OPs: *Send*, *Recv*, *Enqueue*, *Dequeue*.
@@ -48,8 +47,8 @@ After converting:
 
 <img src="src/dist-graph.png" width="700"/>
 
-1. The parameter variable W and it's optimizer subgraph are placed on the parameter server.
-1. Operators are added to the subgraphs.
+1. The parameter variable W and it's optimizer program are placed on the parameter server.
+1. Operators are added to the program.
    - *Send* sends data to the connected *Recv* operator.  The
 	 scheduler on the receive node will only schedule *Recv* operator
 	 to run when the *Send* operator has ran (the *Send* OP will mark
@@ -64,39 +63,30 @@ After converting:
 ### Benefits
 
 - Model parallelism become easier to implement: it's an extension to
-  the trainer - parameter server approach. we already have the
-  communication OPs, but need to extend the graph converter's
-  placement functionality.
-
+  the trainer - parameter server approach. We can have several "Transpilers"
+  to achieve different goals.
 - User-defined optimizer is easier to add - user can now express it as
-  a subgraph.
-
+  a sub-program.
 - No more duplication logic inside the trainer and the parameter
   server mentioned in the background section.
 
 ### Challenges
 
-- It might be hard for the graph converter to cut a general graph
-  (without any hint for which subgraph is the optimizer). We may need
-  to label which subgraph inside the OP graph is the optimizer.
-
 - It's important to balance the parameter shards of on multiple
   parameter server. If a single parameter is very big (some
   word-embedding, fully connected, softmax layer), we need to
   automatically partition the single parameter onto different
   parameter servers when possible (only element-wise optimizer depends
   on the parameter variable).
+- In the "Aync SGD" figure, the "W" variable on the parameter server
+  could be read and wrote concurrently. See
+  [here](https://github.com/PaddlePaddle/Paddle/pull/6394) for more
+  details about concurrent program in fluid.
 
 ### Discussion
 
-- In the "Aync SGD" figure, the "W" variable on the parameter server
-  could be read and wrote concurrently, what is our locking strategy?
-  E.g., each variable have a lock cpp method to be invoked by every
-  OP, or, have a lock OP.
-
 - Can the Enqueue OP be implemented under our current tensor design
   (puts the input tensor into the queue tensor)?
-
 - *Dequeue* OP will have variable numbers of output (depends on the
   `min_count` attribute), does our current design support it? (similar
   question for the *Add* OP)