diff --git a/distributed.py b/distributed.py
index 4e38a6a24f02a3b6d7699dc968ae543140a4ffd6..d4302254738f8354e4e1a41cd58f158f34069249 100644
--- a/distributed.py
+++ b/distributed.py
@@ -22,7 +22,7 @@ import numpy as np
 
 from paddle import fluid
 from paddle.fluid.layers import collective
-from paddle.fluid.dygraph.parallel import Env, ParallelStrategy
+from paddle.fluid.dygraph.parallel import ParallelEnv, ParallelStrategy
 from paddle.fluid.io import BatchSampler
 
 
@@ -64,8 +64,8 @@ class DistributedBatchSampler(BatchSampler):
                 "drop_last should be a boolean number"
 
         self.drop_last = drop_last
-        self.nranks = Env().nranks
-        self.local_rank = Env().local_rank
+        self.nranks = ParallelEnv().nranks
+        self.local_rank = ParallelEnv().local_rank
         self.epoch = 0
         self.num_samples = int(math.ceil(len(self.dataset) * 1.0 / self.nranks))
         self.total_size = self.num_samples * self.nranks
@@ -164,14 +164,14 @@ def init_communicator(program, rank, nranks, wait_port,
 
 def prepare_distributed_context(place=None):
     if place is None:
-        place = fluid.CUDAPlace(Env().dev_id) if Env().nranks > 1 \
+        place = fluid.CUDAPlace(ParallelEnv().dev_id) if ParallelEnv().nranks > 1 \
             else fluid.CUDAPlace(0)
     
     strategy = ParallelStrategy()
-    strategy.nranks = Env().nranks
-    strategy.local_rank = Env().local_rank
-    strategy.trainer_endpoints = Env().trainer_endpoints
-    strategy.current_endpoint = Env().current_endpoint
+    strategy.nranks = ParallelEnv().nranks
+    strategy.local_rank = ParallelEnv().local_rank
+    strategy.trainer_endpoints = ParallelEnv().trainer_endpoints
+    strategy.current_endpoint = ParallelEnv().current_endpoint
 
     if strategy.nranks < 2:
         return