Add cachestat scenario

JIRA: YARDSTICK-267

This scenario reads system cache hit/miss ratio and other statistics using cachestat tool.

Change-Id: I601cb7e64d234571c0a0fd65a5e32604b3a617ec
Signed-off-by: JingLu5 <lvjing5@huawei.com>

6 files changed, 319 insertions, 0 deletions

diff --git a/samples/cachestat.yaml b/samples/cachestat.yaml
new file mode 100644
index 000000000..5786efa38
--- /dev/null
+++ b/samples/cachestat.yaml
@@ -0,0 +1,31 @@
+---
+# Sample benchmark task config file
+# Reading cache hit/miss ratio and usage statistics
+
+schema: "yardstick:task:0.1"
+
+scenarios:
+-
+  type: CACHEstat
+  options:
+    interval: 1
+
+  host: kratos.demo
+
+  runner:
+    type: Duration
+    duration: 60
+
+context:
+  name: demo
+  image: yardstick-trusty-server
+  flavor: yardstick-flavor
+  user: ubuntu
+
+  servers:
+    kratos:
+      floating_ip: true
+
+  networks:
+    test:
+      cidr: '10.0.1.0/24'
diff --git a/tests/unit/benchmark/scenarios/compute/cachestat_sample_output.txt b/tests/unit/benchmark/scenarios/compute/cachestat_sample_output.txt
new file mode 100644
index 000000000..e2c79a9b1
--- /dev/null
+++ b/tests/unit/benchmark/scenarios/compute/cachestat_sample_output.txt
@@ -0,0 +1,5 @@
+Counting cache functions... Output every 1 seconds.
+    HITS   MISSES  DIRTIES    RATIO   BUFFERS_MB   CACHE_MB
+    6462        0       29   100.0%         1157      66782
+
+Ending tracing...
diff --git a/tests/unit/benchmark/scenarios/compute/test_cachestat.py b/tests/unit/benchmark/scenarios/compute/test_cachestat.py
new file mode 100644
index 000000000..f5a6b5ff9
--- /dev/null
+++ b/tests/unit/benchmark/scenarios/compute/test_cachestat.py
@@ -0,0 +1,88 @@
+#!/usr/bin/env python
+
+##############################################################################
+# Copyright (c) 2016 Huawei Technologies Co.,Ltd and others.
+#
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Apache License, Version 2.0
+# which accompanies this distribution, and is available at
+# http://www.apache.org/licenses/LICENSE-2.0
+##############################################################################
+
+# Unittest for yardstick.benchmark.scenarios.compute.cachestat.CACHEstat
+
+import mock
+import unittest
+import os
+
+from yardstick.benchmark.scenarios.compute import cachestat
+
+
+@mock.patch('yardstick.benchmark.scenarios.compute.cachestat.ssh')
+class CACHEstatTestCase(unittest.TestCase):
+
+    def setUp(self):
+        self.ctx = {
+            'host': {
+                'ip': '172.16.0.137',
+                'user': 'root',
+                'key_filename': "mykey.key"
+            }
+        }
+
+        self.result = {}
+
+    def test_cachestat_successful_setup(self, mock_ssh):
+        c = cachestat.CACHEstat({}, self.ctx)
+        mock_ssh.SSH().execute.return_value = (0, '', '')
+
+        c.setup()
+        self.assertIsNotNone(c.client)
+        self.assertTrue(c.setup_done)
+
+    def test_execute_command_success(self, mock_ssh):
+        c = cachestat.CACHEstat({}, self.ctx)
+        mock_ssh.SSH().execute.return_value = (0, '', '')
+        c.setup()
+
+        expected_result = 'abcdefg'
+        mock_ssh.SSH().execute.return_value = (0, expected_result, '')
+        result = c._execute_command("foo")
+        self.assertEqual(result, expected_result)
+
+    def test_execute_command_failed(self, mock_ssh):
+        c = cachestat.CACHEstat({}, self.ctx)
+        mock_ssh.SSH().execute.return_value = (0, '', '')
+        c.setup()
+
+        mock_ssh.SSH().execute.return_value = (127, '', 'Failed executing \
+            command')
+        self.assertRaises(RuntimeError, c._execute_command,
+                          "cat /proc/meminfo")
+
+    def test_get_cache_usage_successful(self, mock_ssh):
+        options = {
+            "interval": 1,
+        }
+        args = {"options": options}
+        c = cachestat.CACHEstat(args, self.ctx)
+        mock_ssh.SSH().execute.return_value = (0, '', '')
+        c.setup()
+
+        output = self._read_file("cachestat_sample_output.txt")
+        mock_ssh.SSH().execute.return_value = (0, output, '')
+        result = c._get_cache_usage()
+        expected_result = {"cachestat": {"cache0": {"HITS": "6462",\
+ "DIRTIES": "29", "RATIO": "100.0%", "MISSES": "0", "BUFFERS_MB": "1157",\
+ "CACHE_MB": "66782"}}, "average": {"HITS": 6462, "DIRTIES": 29, "RATIO": "100.0%",\
+ "MISSES": 0, "BUFFERS_MB":1157, "CACHE_MB": 66782}, "max": {"HITS": 6462,\
+ "DIRTIES": 29, "RATIO": 100.0, "MISSES": 0, "BUFFERS_MB": 1157, "CACHE_MB": 66782}}
+
+        self.assertEqual(result, expected_result)
+
+    def _read_file(self, filename):
+        curr_path = os.path.dirname(os.path.abspath(__file__))
+        output = os.path.join(curr_path, filename)
+        with open(output) as f:
+            sample_output = f.read()
+        return sample_output
diff --git a/tools/ubuntu-server-cloudimg-modify.sh b/tools/ubuntu-server-cloudimg-modify.sh
index f9e0a2c47..5a7b03cef 100755
--- a/tools/ubuntu-server-cloudimg-modify.sh
+++ b/tools/ubuntu-server-cloudimg-modify.sh
@@ -69,5 +69,7 @@ cd /opt/tempT2/ramspeed-2.6.0
 mkdir temp
 bash build.sh
 
+git clone https://github.com/beefyamoeba5/cachestat.git /opt/tempT/Cachestat
+
 # restore symlink
 ln -sf /run/resolvconf/resolv.conf /etc/resolv.conf
diff --git a/yardstick/benchmark/scenarios/compute/cache_stat.bash b/yardstick/benchmark/scenarios/compute/cache_stat.bash
new file mode 100644
index 000000000..393af410c
--- /dev/null
+++ b/yardstick/benchmark/scenarios/compute/cache_stat.bash
@@ -0,0 +1,30 @@
+#!/bin/bash
+
+##############################################################################
+# Copyright (c) 2016 Huawei Technologies Co.,Ltd and others.
+#
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Apache License, Version 2.0
+# which accompanies this distribution, and is available at
+# http://www.apache.org/licenses/LICENSE-2.0
+##############################################################################
+
+# Run a cachestat cache benchmark in a host
+
+set -e
+
+INTERVAL=$1
+
+run_cachestat()
+{
+        cd /opt/tempT/Cachestat
+        bash cachestat $INTERVAL
+}
+
+main()
+{
+    # run the test
+    run_cachestat
+}
+
+main
diff --git a/yardstick/benchmark/scenarios/compute/cachestat.py b/yardstick/benchmark/scenarios/compute/cachestat.py
new file mode 100644
index 000000000..da4aa754f
--- /dev/null
+++ b/yardstick/benchmark/scenarios/compute/cachestat.py
@@ -0,0 +1,163 @@
+##############################################################################
+# Copyright (c) 2016 Huawei Technologies Co.,Ltd and others.
+#
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Apache License, Version 2.0
+# which accompanies this distribution, and is available at
+# http://www.apache.org/licenses/LICENSE-2.0
+##############################################################################
+
+"""cache hit/miss ratio and usage statistics"""
+
+import pkg_resources
+import logging
+import re
+import yardstick.ssh as ssh
+
+from yardstick.benchmark.scenarios import base
+
+LOG = logging.getLogger(__name__)
+
+
+class CACHEstat(base.Scenario):
+    '''Collect cache statistics.
+
+    This scenario reads system cache hit/miss ratio and other statistics on
+    a Linux host.
+
+    cache statistics are read using 'cachestat'.
+    cachestat - show Linux page cache hit/miss statistics.
+                Uses Linux ftrace.
+
+    This is a proof of concept using Linux ftrace capabilities on older
+    kernels, and works by using function profiling for in-kernel counters.
+    Specifically, four kernel functions are traced:
+
+    mark_page_accessed() for measuring cache accesses
+    mark_buffer_dirty() for measuring cache writes
+    add_to_page_cache_lru() for measuring page additions
+    account_page_dirtied() for measuring page dirties
+
+    It is possible that these functions have been renamed (or are different
+    logically) for your kernel version, and this script will not work as-is.
+    This script was written on Linux 3.13. This script is a sandcastle: the
+    kernel may wash some away, and you'll need to rebuild.
+
+    USAGE: cachestat [-Dht] [interval]
+       eg,
+         cachestat 5    # show stats every 5 seconds
+
+    Run "cachestat -h" for full usage.
+
+    WARNING: This uses dynamic tracing of kernel functions, and could cause
+    kernel panics or freezes. Test, and know what you are doing, before use.
+    It also traces cache activity, which can be frequent, and cost some
+    overhead. The statistics should be treated as best-effort: there may be
+    some error margin depending on unusual workload types.
+
+    REQUIREMENTS: CONFIG_FUNCTION_PROFILER, awk.
+    '''
+    __scenario_type__ = "CACHEstat"
+
+    TARGET_SCRIPT = "cache_stat.bash"
+
+    def __init__(self, scenario_cfg, context_cfg):
+        """Scenario construction."""
+        self.scenario_cfg = scenario_cfg
+        self.context_cfg = context_cfg
+        self.setup_done = False
+
+    def setup(self):
+        """Scenario setup."""
+        self.target_script = pkg_resources.resource_filename(
+            "yardstick.benchmark.scenarios.compute",
+            CACHEstat.TARGET_SCRIPT)
+
+        host = self.context_cfg['host']
+        user = host.get('user', 'ubuntu')
+        ip = host.get('ip', None)
+        key_filename = host.get('key_filename', '~/.ssh/id_rsa')
+
+        LOG.info("user:%s, host:%s", user, ip)
+        self.client = ssh.SSH(user, ip, key_filename=key_filename)
+        self.client.wait(timeout=600)
+
+        # copy scripts to host
+        self.client.run("cat > ~/cache_stat.sh",
+                        stdin=open(self.target_script, 'rb'))
+
+        self.setup_done = True
+
+    def _execute_command(self, cmd):
+        """Execute a command on server."""
+        LOG.info("Executing: %s" % cmd)
+        status, stdout, stderr = self.client.execute(cmd)
+        if status:
+            raise RuntimeError("Failed executing command: ",
+                               cmd, stderr)
+        return stdout
+
+    def _filtrate_result(self, result):
+        fields = []
+        cachestat = {}
+        data_marker = re.compile("\d+")
+        ite = 0
+        average = {'HITS': 0, 'MISSES': 0, 'DIRTIES': 0, 'RATIO': 0,
+                   'BUFFERS_MB': 0, 'CACHE_MB': 0}
+        maximum = {'HITS': 0, 'MISSES': 0, 'DIRTIES': 0, 'RATIO': 0,
+                   'BUFFERS_MB': 0, 'CACHE_MB': 0}
+
+        # Parse cache stats
+        for row in result.split('\n'):
+            line = row.split()
+
+            if line and line[0] == 'HITS':
+                # header fields
+                fields = line[:]
+            elif line and re.match(data_marker, line[0]):
+                cache = 'cache' + str(ite)
+                ite += 1
+                values = line[:]
+                if values and len(values) == len(fields):
+                    cachestat[cache] = dict(zip(fields, values))
+
+        for entry in cachestat:
+            for item in average:
+                if item != 'RATIO':
+                    average[item] += int(cachestat[entry][item])
+                else:
+                    average[item] += float(cachestat[entry][item][:-1])
+
+            for item in maximum:
+                if item != 'RATIO':
+                    if int(cachestat[entry][item]) > maximum[item]:
+                        maximum[item] = int(cachestat[entry][item])
+                else:
+                    if float(cachestat[entry][item][:-1]) > maximum[item]:
+                        maximum[item] = float(cachestat[entry][item][:-1])
+
+        for item in average:
+            if item != 'RATIO':
+                average[item] = average[item] / len(cachestat)
+            else:
+                average[item] = str(average[item] / len(cachestat)) + '%'
+
+        return {'cachestat': cachestat, 'average': average, 'max': maximum}
+
+    def _get_cache_usage(self):
+        """Get cache statistics."""
+        options = self.scenario_cfg['options']
+        interval = options.get("interval", 1)
+
+        cmd = "sudo bash cache_stat.sh %s" % (interval)
+
+        result = self._execute_command(cmd)
+        filtrated_result = self._filtrate_result(result)
+
+        return filtrated_result
+
+    def run(self, result):
+        if not self.setup_done:
+            self.setup()
+
+        result.update(self._get_cache_usage())