diff --git a/benchmark/benchmark.py b/benchmark/benchmark.py
new file mode 100644
index 0000000..f671500
--- /dev/null
+++ b/benchmark/benchmark.py
@@ -0,0 +1,101 @@
+# Copyright (c) 2014 Mirantis Inc.
+# 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.
+
+from __future__ import print_function
+
+import atexit
+import math
+import os
+import subprocess
+import sys
+import timeit
+
+config_path = "rootwrap.conf"
+num_iterations = 100
+
+
+def run_plain(cmd):
+    obj = subprocess.Popen(cmd,
+                           stdin=subprocess.PIPE,
+                           stdout=subprocess.PIPE,
+                           stderr=subprocess.PIPE)
+    out, err = obj.communicate()
+    return obj.returncode, out, err
+
+
+def run_sudo(cmd):
+    return run_plain(["sudo"] + cmd)
+
+
+def run_rootwrap(cmd):
+    return run_plain([
+        "sudo", sys.executable, "-c",
+        "from oslo.rootwrap import cmd; cmd.main()", config_path] + cmd)
+
+
+def run_one(runner, cmd):
+    def __inner():
+        code, out, err = runner(cmd)
+        assert err == "", "Stderr not empty:\n" + err
+        assert code == 0, "Command failed"
+    return __inner
+
+runners = [
+    ("{0}", run_plain),
+    ("sudo {0}", run_sudo),
+    ("sudo rootwrap conf {0}", run_rootwrap),
+]
+
+
+def get_time_string(sec):
+    if sec > 0.9:
+        return "{0:7.3f}s ".format(sec)
+    elif sec > 0.0009:
+        return "{0:7.3f}ms".format(sec * 1000.0)
+    else:
+        return "{0:7.3f}us".format(sec * 1000000.0)
+
+
+def run_bench(cmd, runners):
+    strcmd = ' '.join(cmd)
+    max_name_len = max(len(name) for name, _ in runners) + len(strcmd) - 3
+    print("Running '{0}':".format(strcmd))
+    print("{0:^{1}} :".format("method", max_name_len),
+          "".join(map("{0:^10}".format, ["min", "avg", "max", "dev"])))
+    for name, runner in runners:
+        results = timeit.repeat(run_one(runner, cmd), repeat=num_iterations,
+                                number=1)
+        avg = sum(results) / num_iterations
+        min_ = min(results)
+        max_ = max(results)
+        dev = math.sqrt(sum((r - avg) ** 2 for r in results) / num_iterations)
+        print("{0:>{1}} :".format(name.format(strcmd), max_name_len),
+              " ".join(map(get_time_string, [min_, avg, max_, dev])))
+
+
+def main():
+    os.chdir(os.path.dirname(__file__))
+    code, _, _ = run_sudo(["-vn"])
+    if code:
+        print("We need you to authorize with sudo to run this benchmark")
+        run_sudo(["-v"])
+
+    run_bench(["ip", "a"], runners)
+    run_sudo(["ip", "netns", "add", "bench_ns"])
+    atexit.register(run_sudo, ["ip", "netns", "delete", "bench_ns"])
+    run_bench('ip netns exec bench_ns ip a'.split(), runners[1:])
+
+if __name__ == "__main__":
+    main()
diff --git a/benchmark/filters.d/ip.filters b/benchmark/filters.d/ip.filters
new file mode 100644
index 0000000..8942d6d
--- /dev/null
+++ b/benchmark/filters.d/ip.filters
@@ -0,0 +1,4 @@
+[Filters]
+
+ip: IpFilter, ip, root
+ip_exec: IpNetnsExecFilter, ip, root
diff --git a/benchmark/rootwrap.conf b/benchmark/rootwrap.conf
new file mode 100644
index 0000000..73240df
--- /dev/null
+++ b/benchmark/rootwrap.conf
@@ -0,0 +1,4 @@
+[DEFAULT]
+filters_path=filters.d
+exec_dirs=/sbin,/usr/sbin,/bin,/usr/bin
+use_syslog=False
diff --git a/tox.ini b/tox.ini
index ca6cd00..39b563c 100644
--- a/tox.ini
+++ b/tox.ini
@@ -26,3 +26,6 @@ commands = {posargs}
 show-source = True
 exclude = .tox,dist,doc,*.egg,build
 builtins = _
+
+[testenv:benchmark]
+commands = python benchmark/benchmark.py