diff --git a/paddle/framework/CMakeLists.txt b/paddle/framework/CMakeLists.txt
index 2be21e825ae1b028eefe820e4e152a0666d67f10..1afc5242081e7f7b12527a15d29421cebeb3d3b8 100644
--- a/paddle/framework/CMakeLists.txt
+++ b/paddle/framework/CMakeLists.txt
@@ -45,8 +45,9 @@ add_custom_command(TARGET framework_py_proto POST_BUILD
 
 cc_library(backward SRCS backward.cc DEPS net_op)
 cc_test(backward_test SRCS backward_test.cc DEPS backward recurrent_op device_context fill_constant_op)
+cc_library(lod_rank_table SRCS lod_rank_table.cc DEPS lod_tensor)
 
-cc_library(executor SRCS executor.cc DEPS op_registry device_context scope framework_proto backward glog)
+cc_library(executor SRCS executor.cc DEPS op_registry device_context scope framework_proto backward glog lod_rank_table)
 
 cc_library(prune SRCS prune.cc DEPS framework_proto)
 cc_test(prune_test SRCS prune_test.cc DEPS op_info prune recurrent_op device_context)
diff --git a/paddle/framework/executor.cc b/paddle/framework/executor.cc
index 52fefe4ea30899880cd386587340d691ee97547b..c1a009f131c803f7881faf2e322eecf3e6e8ea05 100644
--- a/paddle/framework/executor.cc
+++ b/paddle/framework/executor.cc
@@ -21,6 +21,7 @@ limitations under the License. */
 #include <vector>
 
 #include "paddle/framework/feed_fetch_type.h"
+#include "paddle/framework/lod_rank_table.h"
 #include "paddle/framework/lod_tensor.h"
 #include "paddle/framework/op_registry.h"
 #include "paddle/framework/scope.h"
@@ -70,10 +71,12 @@ static void CreateTensor(Variable* var, VarDesc::VarType var_type) {
     var->GetMutable<FeedFetchList>();
   } else if (var_type == VarDesc::STEP_SCOPES) {
     var->GetMutable<std::vector<framework::Scope>>();
+  } else if (var_type == VarDesc::LOD_RANK_TABLE) {
+    var->GetMutable<LoDRankTable>();
   } else {
     PADDLE_THROW(
         "Variable type %d is not in "
-        "[LoDTensor, SelectedRows, FEED_MINIBATCH, FETCH_LIST]",
+        "[LoDTensor, SelectedRows, FEED_MINIBATCH, FETCH_LIST, LOD_RANK_TABLE]",
         var_type);
   }
 }
diff --git a/paddle/framework/framework.proto b/paddle/framework/framework.proto
index 8f2df3dc0e29f96b3aea58b6761d1ccb4cd7c624..54ce461ce81955c9c3b550998e4a1df32c201ab3 100644
--- a/paddle/framework/framework.proto
+++ b/paddle/framework/framework.proto
@@ -116,6 +116,7 @@ message VarDesc {
     FEED_MINIBATCH = 3;
     FETCH_LIST = 4;
     STEP_SCOPES = 5;
+    LOD_RANK_TABLE = 6;
   }
   required string name = 1;
   required VarType type = 2;
diff --git a/paddle/framework/lod_rank_table.cc b/paddle/framework/lod_rank_table.cc
new file mode 100644
index 0000000000000000000000000000000000000000..f9abf902a13f3ebe5ea75c7456b117462cec8da5
--- /dev/null
+++ b/paddle/framework/lod_rank_table.cc
@@ -0,0 +1,43 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+   http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License. */
+
+#include "paddle/framework/lod_rank_table.h"
+
+namespace paddle {
+namespace framework {
+void LoDRankTable::Reset(const LoD& lod, size_t level) {
+  this->coarse_lod_.clear();
+  this->items_.clear();
+  PADDLE_ENFORCE(level < lod.size(),
+                 "Cannot rank lod since the level %d is less than lod size %d",
+                 level, lod.size());
+  coarse_lod_.reserve(level);
+  for (size_t i = 0; i < level; ++i) {
+    coarse_lod_.push_back(lod[i]);
+  }
+  auto& vec = lod[level];
+  for (size_t i = 0; i < vec.size() - 1; ++i) {
+    TableItem item;
+    item.index = i;
+    item.length = vec[i + 1] - vec[i];
+    items_.emplace_back(item);
+  }
+  std::sort(items_.begin(), items_.end(),
+            [](const TableItem& a, const TableItem& b) {
+              return a.length > b.length;
+            });
+}
+
+}  // namespace framework
+}  // namespace paddle
diff --git a/paddle/framework/lod_rank_table.h b/paddle/framework/lod_rank_table.h
new file mode 100644
index 0000000000000000000000000000000000000000..9faa3a4d7bdc55ab7b24e31f5e5434dacc0a4b36
--- /dev/null
+++ b/paddle/framework/lod_rank_table.h
@@ -0,0 +1,55 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+   http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License. */
+
+#pragma once
+#include "paddle/framework/lod_tensor.h"
+
+namespace paddle {
+namespace framework {
+
+// LoD Rank Table stores the `level` of `lod` which is ordered by sequence
+// length in descending order. It is useful when implement dynamic RNN and is
+// shared by dynamic RNN memory, dynamic RNN slice input and dynamic RNN slice
+// output operators.
+//
+// The table item contains two element. The length of sequence and the index of
+// sequence in that level.
+//
+// LoDRankTable also stores the coarse_lod, which is the lod information whose
+// level is less than input level, in order to restore the output LoD
+// information.
+class LoDRankTable {
+ public:
+  struct TableItem {
+    size_t index;
+    size_t length;
+  };
+
+  LoDRankTable() {}
+
+  void Reset(const LoD& lod, size_t level);
+
+  const std::vector<TableItem>& items() const { return this->items_; }
+
+  const LoD& coarse_lod() const { return this->coarse_lod_; }
+
+  size_t level() const { return coarse_lod_.size(); }
+
+ private:
+  LoD coarse_lod_;
+  std::vector<TableItem> items_;
+};
+
+}  // namespace framework
+}  // namespace paddle
diff --git a/paddle/framework/var_desc.h b/paddle/framework/var_desc.h
index 70daa20e8d99abc5759655adf538a8c197e9ec6a..5cf4608944c5011d798fbde060002a57be8f6102 100644
--- a/paddle/framework/var_desc.h
+++ b/paddle/framework/var_desc.h
@@ -15,6 +15,7 @@ limitations under the License. */
 #pragma once
 
 #include <vector>
+#include "glog/logging.h"
 #include "paddle/framework/framework.pb.h"
 
 namespace paddle {
diff --git a/paddle/operators/CMakeLists.txt b/paddle/operators/CMakeLists.txt
index 81d92ec6f4f8c94e08d3b86b6319a9bf06f76a22..13ebb0ad65f62b5961f3280bf6b178be5c8fe30b 100644
--- a/paddle/operators/CMakeLists.txt
+++ b/paddle/operators/CMakeLists.txt
@@ -141,6 +141,7 @@ set(DEPS_OPS
     pool_with_index_op
     nccl_op
     sequence_conv_op
+    lod_rank_table_op
     lstm_op)
 
 op_library(cond_op SRCS cond_op.cc DEPS framework_proto tensor operator net_op)
@@ -149,6 +150,7 @@ op_library(softmax_with_cross_entropy_op DEPS cross_entropy softmax)
 op_library(sum_op DEPS net_op selected_rows_functor)
 op_library(pool_op DEPS pooling)
 op_library(pool_with_index_op DEPS pooling)
+op_library(lod_rank_table_op SRCS lod_rank_table_op.cc DEPS lod_rank_table)
 if(WITH_GPU)
 op_library(nccl_op DEPS nccl_common)
 endif()
diff --git a/paddle/operators/lod_rank_table_op.cc b/paddle/operators/lod_rank_table_op.cc
new file mode 100644
index 0000000000000000000000000000000000000000..be198951c241dc5e5587c8a2b8d94f67173d2b2a
--- /dev/null
+++ b/paddle/operators/lod_rank_table_op.cc
@@ -0,0 +1,80 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+   http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License. */
+#include "paddle/framework/lod_rank_table.h"
+#include "paddle/framework/op_registry.h"
+namespace paddle {
+namespace operators {
+
+class LoDRankTableOp : public framework::OperatorBase {
+ public:
+  LoDRankTableOp(const std::string &type,
+                 const framework::VariableNameMap &inputs,
+                 const framework::VariableNameMap &outputs,
+                 const framework::AttributeMap &attrs)
+      : OperatorBase(type, inputs, outputs, attrs) {}
+  void Run(const framework::Scope &scope,
+           const platform::DeviceContext &dev_ctx) const override {
+    auto x = scope.FindVar(Input("X"))->Get<framework::LoDTensor>();
+    auto *out =
+        scope.FindVar(Output("Out"))->GetMutable<framework::LoDRankTable>();
+    out->Reset(x.lod(), static_cast<size_t>(Attr<int>("level")));
+  }
+};
+
+class LoDRankTableOpProtoMaker : public framework::OpProtoAndCheckerMaker {
+ public:
+  LoDRankTableOpProtoMaker(framework::OpProto *proto,
+                           framework::OpAttrChecker *op_checker)
+      : OpProtoAndCheckerMaker(proto, op_checker) {
+    AddInput("X",
+             "(LoDTensor) input lod tensor, must contain lod information.");
+    AddOutput("Out", "(LoDRankTable) The rank table of specific level.");
+    AddAttr<int>("level", "(int) the specific lod level to rank.")
+        .SetDefault(0)
+        .EqualGreaterThan(0);
+    AddComment(R"DOC(Create LoDRanTable by LoDTensor
+
+LoD Rank Table stores the `level` of `lod` which is ordered by sequence
+length in descending order. It is useful when implement dynamic RNN and is
+shared by dynamic RNN memory, dynamic RNN slice input and dynamic RNN slice
+output operators.
+)DOC");
+  }
+};
+
+class LoDRankTableInferShape : public framework::InferShapeBase {
+ public:
+  void operator()(framework::InferShapeContext *context) const override {
+    PADDLE_ENFORCE(context->HasInput("X"), "LoDRankTable must has input X");
+  }
+};
+
+class LoDRankTableInferVarType : public framework::VarTypeInference {
+ public:
+  void operator()(const framework::OpDescBind &op_desc,
+                  framework::BlockDescBind *block) const override {
+    for (auto &o : op_desc.Output("Out")) {
+      block->Var(o)->SetType(framework::VarDesc::LOD_RANK_TABLE);
+    }
+  }
+};
+
+}  // namespace operators
+}  // namespace paddle
+
+REGISTER_OPERATOR(lod_rank_table, paddle::operators::LoDRankTableOp,
+                  paddle::operators::LoDRankTableOpProtoMaker,
+                  paddle::operators::LoDRankTableInferShape,
+                  paddle::operators::LoDRankTableInferVarType,
+                  paddle::framework::EmptyGradOpMaker);
diff --git a/paddle/pybind/protobuf.cc b/paddle/pybind/protobuf.cc
index dcae426c7e231757d796c5a84cc5a1c2b0d6763b..d3fc544ec712a753304de8568534c78c154774a5 100644
--- a/paddle/pybind/protobuf.cc
+++ b/paddle/pybind/protobuf.cc
@@ -238,7 +238,8 @@ void BindVarDsec(py::module &m) {
       .value("SELECTED_ROWS", VarDesc::SELECTED_ROWS)
       .value("FEED_MINIBATCH", VarDesc::FEED_MINIBATCH)
       .value("FETCH_LIST", VarDesc::FETCH_LIST)
-      .value("STEP_SCOPES", VarDesc::STEP_SCOPES);
+      .value("STEP_SCOPES", VarDesc::STEP_SCOPES)
+      .value("LOD_RANK_TABLE", VarDesc::LOD_RANK_TABLE);
 }
 
 void BindOpDesc(py::module &m) {
diff --git a/paddle/pybind/pybind.cc b/paddle/pybind/pybind.cc
index aab08a759b094382c62feec57c6a907490331fea..78dc7943b3644e7014da427e6aca640d94746f68 100644
--- a/paddle/pybind/pybind.cc
+++ b/paddle/pybind/pybind.cc
@@ -21,6 +21,7 @@ limitations under the License. */
 #include "paddle/framework/executor.h"
 #include "paddle/framework/feed_fetch_method.h"
 #include "paddle/framework/framework.pb.h"
+#include "paddle/framework/lod_rank_table.h"
 #include "paddle/framework/lod_tensor.h"
 #include "paddle/framework/prune.h"
 #include "paddle/framework/selected_rows.h"
@@ -224,6 +225,9 @@ All parameter, weight, gradient are variables in Paddle.
              return self.GetMutable<LoDTensor>();
            },
            py::return_value_policy::reference)
+      .def("get_lod_rank_table",
+           [](Variable &self) { return self.GetMutable<LoDRankTable>(); },
+           py::return_value_policy::reference)
       .def("get_selected_rows",
            [](Variable &self) -> SelectedRows * {
              return self.GetMutable<SelectedRows>();
@@ -492,6 +496,15 @@ All parameter, weight, gradient are variables in Paddle.
   BindVarDsec(m);
   BindOpDesc(m);
 
+  py::class_<framework::LoDRankTable>(m, "LodRankTable")
+      .def("items", [](framework::LoDRankTable &table) {
+        std::vector<std::pair<size_t, size_t>> res;
+        for (auto &item : table.items()) {
+          res.push_back({item.index, item.length});
+        }
+        return res;
+      });
+
   m.def("op_support_gpu", OpSupportGPU);
 #ifdef PADDLE_WITH_CUDA
   m.def("get_cuda_device_count", platform::GetCUDADeviceCount);
diff --git a/python/paddle/v2/framework/framework.py b/python/paddle/v2/framework/framework.py
index a890bbf598569ecb7d8f9f6b02ca4ab9895d149e..4e737549c929c6d5b2f2784f4fe9bcfeedae66e2 100644
--- a/python/paddle/v2/framework/framework.py
+++ b/python/paddle/v2/framework/framework.py
@@ -101,6 +101,10 @@ class Variable(object):
     def persistable(self):
         return self.desc.persistable()
 
+    @persistable.setter
+    def persistable(self, p):
+        self.desc.set_persistable(p)
+
     @property
     def name(self):
         return self.desc.name()
diff --git a/python/paddle/v2/framework/layers.py b/python/paddle/v2/framework/layers.py
index a98b4e554f9877436381ced6a2576bbe286feb3f..d6b5be945825131a7bdefa4f972a690543f2a9b6 100644
--- a/python/paddle/v2/framework/layers.py
+++ b/python/paddle/v2/framework/layers.py
@@ -729,3 +729,16 @@ class StaticRNN(object):
                 'states': memories,
                 'step_block': rnn_block
             })
+
+
+def lod_rank_table(x, level=0, program=None):
+    helper = LayerHelper("lod_rank_table", **locals())
+    table = helper.create_variable(
+        type=core.VarDesc.VarType.LOD_RANK_TABLE,
+        name=unique_name("lod_rank_table"))
+    helper.append_op(
+        type='lod_rank_table',
+        inputs={'X': x},
+        outputs={'Out': table},
+        attrs={'level': level})
+    return table
diff --git a/python/paddle/v2/framework/tests/test_lod_rank_table.py b/python/paddle/v2/framework/tests/test_lod_rank_table.py
new file mode 100644
index 0000000000000000000000000000000000000000..f635e716bcd0b338159235c1c66c490a14d53b07
--- /dev/null
+++ b/python/paddle/v2/framework/tests/test_lod_rank_table.py
@@ -0,0 +1,29 @@
+from paddle.v2.framework.layers import lod_rank_table, data
+from paddle.v2.framework.executor import Executor
+from paddle.v2.framework.framework import g_program
+import paddle.v2.framework.core as core
+import numpy
+import unittest
+
+
+class TestLoDRankTable(unittest.TestCase):
+    def test_lod_rank_table(self):
+        x = data(name='x', shape=[100])
+        cpu = core.CPUPlace()
+        rank_table = lod_rank_table(x=x, level=1)
+        rank_table.persistable = True
+        exe = Executor(cpu)
+        scope = core.Scope()
+
+        tensor = core.LoDTensor()
+        tensor.set(numpy.random.random(size=(17, 100)), cpu)
+        tensor.set_lod([[0, 1, 3], [0, 5, 6, 7], [0, 3, 4, 9, 10, 13, 16, 17]])
+
+        exe.run(g_program, scope=scope, feed={'x': tensor})
+        var = scope.find_var(rank_table.name)
+        table = var.get_lod_rank_table()
+        self.assertEqual([(0, 5), (1, 1), (2, 1)], table.items())
+
+
+if __name__ == '__main__':
+    unittest.main()