distribute transpiler support async config

Author: jacquesqiao

python/paddle/fluid/distribute_transpiler.py

@@ -358,7 +358,7 @@ def get_pserver_program(self, endpoint):
                     type=v.type,
                     dtype=v.dtype,
                     shape=v.shape)
-            if self.trainer_num > 1:
+            if self.sync_mode and self.trainer_num > 1:
                 for trainer_id in xrange(self.trainer_num):
                     var = pserver_program.global_block().create_var(
                         name="%s.trainer_%d" % (orig_var_name, trainer_id),
@@ -688,17 +688,6 @@ def _clone_var(block, var, persistable=True):
                 self.table_name)],
             persistable=False)
 
-        # create grad vars in pserver program
-        table_grad_var = self.table_param_grad[1]
-        table_grad_list = [
-            pserver_program.global_block().create_var(
-                name="%s.trainer_%d.pserver_%d" %
-                (table_grad_var.name, index, pserver_index),
-                type=table_grad_var.type,
-                shape=table_grad_var.shape,
-                dtype=table_grad_var.dtype) for index in range(self.trainer_num)
-        ]
-
         # create table optimize block in pserver program
         table_opt_op = [
             op for op in self.optimize_ops
@@ -708,11 +697,24 @@ def _clone_var(block, var, persistable=True):
         # only support sgd now
         assert table_opt_op.type == "sgd"
 
-        # append sum op for table_grad_list
-        table_opt_block.append_op(
-            type="sum",
-            inputs={"X": table_grad_list},
-            outputs={"Out": [grad_var]})
+        if self.sync_mode:
+            # create grad vars in pserver program
+            table_grad_var = self.table_param_grad[1]
+            table_grad_list = [
+                pserver_program.global_block().create_var(
+                    name="%s.trainer_%d.pserver_%d" %
+                    (table_grad_var.name, index, pserver_index),
+                    type=table_grad_var.type,
+                    shape=table_grad_var.shape,
+                    dtype=table_grad_var.dtype)
+                for index in range(self.trainer_num)
+            ]
+
+            # append sum op for table_grad_list
+            table_opt_block.append_op(
+                type="sum",
+                inputs={"X": table_grad_list},
+                outputs={"Out": [grad_var]})
 
         lr_var = pserver_program.global_block().vars[table_opt_op.input(
             "LearningRate")[0]]
@@ -751,7 +753,7 @@ def _create_vars_from_blocklist(self,
         for varname, splited in block_map.iteritems():
             orig_var = program.global_block().var(varname)
             if len(splited) == 1:
-                if add_trainer_suffix:
+                if self.sync_mode and add_trainer_suffix:
                     new_var_name = "%s.trainer_%d" % \
                         (orig_var.name, self.trainer_id)
                     program.global_block().rename_var(varname, new_var_name)
@@ -775,7 +777,7 @@ def _create_vars_from_blocklist(self,
                 if len(orig_shape) >= 2:
                     splited_shape.extend(orig_shape[1:])
                 new_var_name = ""
-                if add_trainer_suffix:
+                if self.sync_mode and add_trainer_suffix:
                     new_var_name = "%s.block%d.trainer_%d" % \
                         (varname, i, self.trainer_id)
                 else:
@@ -907,7 +909,7 @@ def _append_pserver_ops(self, optimize_block, opt_op, endpoint,
                     pserver_block.vars[self._orig_varname(grad_block.name)]
                 grad_to_block_id.append(merged_var.name + ":" + str(
                     optimize_block.idx))
-                if self.trainer_num > 1:
+                if self.sync_mode and self.trainer_num > 1:
                     vars2merge = []
                     for i in xrange(self.trainer_num):
                         per_trainer_name = "%s.trainer_%d" % \
@@ -925,6 +927,7 @@ def _append_pserver_ops(self, optimize_block, opt_op, endpoint,
                             inputs={"X": merged_var},
                             outputs={"Out": merged_var},
                             attrs={"scale": 1.0 / float(self.trainer_num)})
+
                 new_inputs[key] = merged_var
             elif key == "Param":
                 # param is already created on global program