diff --git a/python/examples/criteo_ctr_with_cube/benchmark.py b/python/examples/criteo_ctr_with_cube/benchmark.py
index df5c6b90badb36fd7e349555973ccbd7ea0a8b70..a850d244b0a5a1a01e98a6207fa9674b6ea0af1a 100755
--- a/python/examples/criteo_ctr_with_cube/benchmark.py
+++ b/python/examples/criteo_ctr_with_cube/benchmark.py
@@ -46,7 +46,7 @@ def single_func(idx, resource):
         start = time.time()
         itr = 1000
         for ei in range(itr):
-            if args.batch_size > 1:
+            if args.batch_size > 0:
                 feed_batch = []
                 for bi in range(args.batch_size):
                     data = reader().next()
diff --git a/python/examples/criteo_ctr_with_cube/benchmark.sh b/python/examples/criteo_ctr_with_cube/benchmark.sh
index a7f927e186dd98e6f04de96d37bee7dd7b62359b..35b19b637d9e8dec10fd3b59224c5c17e3ba5f53 100755
--- a/python/examples/criteo_ctr_with_cube/benchmark.sh
+++ b/python/examples/criteo_ctr_with_cube/benchmark.sh
@@ -1,9 +1,13 @@
 rm profile_log
+export FLAGS_profile_client=1
+export FLAGS_profile_server=1
 for thread_num in 1 2 4 8 16
 do
-for batch_size in 1 2 4 8 16 32 64 128 256 512
+for batch_size in 1 4 16 64 256
 do
     $PYTHONROOT/bin/python benchmark.py --thread $thread_num --batch_size $batch_size --model serving_client_conf/serving_client_conf.prototxt --request rpc > profile 2>&1
+    echo "batch size : $batch_size"
+    echo "thread num : $thread_num"
     echo "========================================"
     echo "batch size : $batch_size" >> profile_log
     $PYTHONROOT/bin/python ../util/show_profile.py profile $thread_num >> profile_log