TechEmpower.FrameworkBenchmarks/benchmarker.py

from installer import Installer
from framework_test import FrameworkTest
import framework_test

import os
import json
import subprocess
import time
import textwrap
import pprint
import csv
import sys
from datetime import datetime

class Benchmarker:

  ##########################################################################################
  # Public methods
  ##########################################################################################

  ############################################################
  # Prints all the available tests
  ############################################################
  def run_list_tests(self):
    all_tests = self.__gather_tests()

    for test in all_tests:
      print test.name

    self.__finish()

  ############################################################
  # End run_list_tests
  ############################################################

  ############################################################
  # next_sort
  # Prints the next available sort number that should be used 
  # for any new tests
  ############################################################
  def next_sort_value(self):
    all_tests = self.__gather_tests()

    # all_tests is already sorted by sort, so we can just get
    # the last one and add one to it.
    print " Next sort number is: " + str(all_tests[-1].sort + 1)

    self.__finish()

  ############################################################
  # End next_sort_value
  ############################################################
  
  ############################################################
  # parse_timestamp
  # Re-parses the raw data for a given timestamp
  ############################################################
  def parse_timestamp(self):
    all_tests = self.__gather_tests()
    
    for test in all_tests:
      test.parse_all()
    
    self.__parse_results(all_tests)

    self.__finish()

  ############################################################
  # End run_list_tests
  ############################################################

  ############################################################
  # Run the tests:
  # This process involves setting up the client/server machines
  # with any necessary change. Then going through each test,
  # running their setup script, verifying the URLs, and
  # running benchmarks against them.
  ############################################################
  def run(self):
    ##########################
    # Get a list of all known
    # tests that we can run.
    ##########################    
    all_tests = self.__gather_tests()

    ##########################
    # Setup client/server
    ##########################
    print textwrap.dedent("""
      =====================================================
        Preparing up Server and Client ...
      =====================================================
      """)
    self.__setup_server()
    self.__setup_client()

    ##########################
    # Run tests
    ##########################
    self.__run_tests(all_tests)

    ##########################
    # Parse results
    ##########################  
    if self.mode == "benchmark":
      print textwrap.dedent("""
      =====================================================
        Parsing Results ...
      =====================================================
      """)
      self.__parse_results(all_tests)

    self.__finish()

  ############################################################
  # End run
  ############################################################

  ############################################################
  # sftp_string(batch_file)
  # generates a fully qualified URL for sftp to client
  ############################################################
  def sftp_string(self, batch_file):
    sftp_string =  "sftp -oStrictHostKeyChecking=no " 
    if batch_file != None: sftp_string += " -b " + batch_file + " "
    
    if self.identity_file != None:
      sftp_string += " -i " + self.identity_file + " "

    return sftp_string + self.client_user + "@" + self.client_host
  ############################################################
  # End sftp_string
  ############################################################

  ############################################################
  # generate_url(url, port)
  # generates a fully qualified URL for accessing a test url
  ############################################################
  def generate_url(self, url, port):
    return self.server_host + ":" + str(port) + url
  ############################################################
  # End generate_url
  ############################################################

  ############################################################
  # output_file(test_name, test_type)
  # returns the output file for this test_name and test_type
  # timestamp/test_type/test_name/raw 
  ############################################################
  def output_file(self, test_name, test_type):
    path = os.path.join(self.result_directory, self.timestamp, test_type, test_name, "raw")
    try:
      os.makedirs(os.path.dirname(path))
    except OSError:
      pass
    return path
  ############################################################
  # End output_file
  ############################################################

  ############################################################
  # full_results_directory
  ############################################################
  def full_results_directory(self):
    path = os.path.join(self.result_directory, self.timestamp)
    try:
      os.makedirs(path)
    except OSError:
      pass
    return path
  ############################################################
  # End output_file
  ############################################################

  ############################################################
  # report_results
  ############################################################
  def report_results(self, framework, test, results, latency, requests, total_time, errors, total_requests):
    # Try to get the id in the result array if it exists.
    try:
      framework_id = str(self.results['frameworks'].index(framework.name))
    except ValueError:
      framework_id = str(framework.sort)
      
    if test not in self.results['rawData'].keys():
      self.results['rawData'][test] = dict()
      self.results['weighttpData'][test] = dict()

    self.results['rawData'][test][framework_id] = results
    self.results['weighttpData'][test][framework_id] = dict()
    self.results['weighttpData'][test][framework_id]['latency'] = latency
    self.results['weighttpData'][test][framework_id]['requests'] = requests
    self.results['weighttpData'][test][framework_id]['totalTime'] = total_time
    self.results['weighttpData'][test][framework_id]['errors'] = errors
    self.results['weighttpData'][test][framework_id]['totalRequests'] = total_requests

  ############################################################
  # End report_results
  ############################################################

  ##########################################################################################
  # Private methods
  ##########################################################################################

  ############################################################
  # Gathers all the tests
  ############################################################
  def __gather_tests(self):
    tests = []
    # Loop through each directory (we assume we're being run from the benchmarking root)
    # and look for the files that signify a benchmark test
    for dirname, dirnames, filenames in os.walk('.'):
      # Look for the benchmark_config file, this will set up our tests
      # It's format looks like this:
      #
      # {
      #   "framework": "nodejs",
      #   "tests": [{
      #     "default": {
      #       "setup_file": "setup",
      #       "json_url": "/json"
      #     },
      #     "mysql": {
      #       "setup_file": "setup",
      #       "db_url": "/mysql",
      #       "query_url": "/mysql?queries="
      #     },
      #     ...
      #   }]
      # }
      if 'benchmark_config' in filenames:
        config = None
        with open(os.path.join(dirname, 'benchmark_config'), 'r') as config_file:
          # Load json file into config object
          config = json.load(config_file)

        if config == None:
          continue

        tests = tests + framework_test.parse_config(config, dirname[2:], self)

    tests.sort(key=lambda x: x.sort)
    return tests
  ############################################################
  # End __gather_tests
  ############################################################

  ############################################################
  # Makes any necessary changes to the server that should be 
  # made before running the tests. This involves setting kernal
  # settings to allow for more connections, or more file
  # descriptiors
  #
  # http://redmine.lighttpd.net/projects/weighttp/wiki#Troubleshooting
  ############################################################
  def __setup_server(self):
    try:
      subprocess.check_call("sudo sysctl -w net.core.somaxconn=1024".rsplit(" "))
      subprocess.check_call("sudo -s ulimit -n 8192".rsplit(" "))
      subprocess.check_call("sudo sysctl net.ipv4.tcp_tw_reuse=1".rsplit(" "))
      subprocess.check_call("sudo sysctl net.ipv4.tcp_tw_recycle=1".rsplit(" "))
      subprocess.check_call("sudo sysctl -w kernel.shmmax=134217728".rsplit(" "))
      subprocess.check_call("sudo sysctl -w kernel.shmall=2097152".rsplit(" "))
    except subprocess.CalledProcessError:
      return False
  ############################################################
  # End __setup_server
  ############################################################

  ############################################################
  # Makes any necessary changes to the client machine that 
  # should be made before running the tests. Is very similar
  # to the server setup, but may also include client specific
  # changes.
  ############################################################
  def __setup_client(self):
    p = subprocess.Popen(self.ssh_string, stdin=subprocess.PIPE, shell=True)
    p.communicate("""
      sudo sysctl -w net.core.somaxconn=1024
      sudo -s ulimit -n 8192
      sudo sysctl net.ipv4.tcp_tw_reuse=1
      sudo sysctl net.ipv4.tcp_tw_recycle=1
      sudo sysctl -w kernel.shmmax=134217728
      sudo sysctl -w kernel.shmall=2097152
    """)
  ############################################################
  # End __setup_client
  ############################################################

  ############################################################
  # __run_tests
  # Ensures that the system has all necessary software to run
  # the tests. This does not include that software for the individual
  # test, but covers software such as curl and weighttp that
  # are needed.
  ############################################################
  def __run_tests(self, tests):
    for test in tests:
      # If the user specified which tests to run, then 
      # we can skip over tests that are not in that list
      if self.test != None and test.name not in self.test:
        continue
      
      # If the test is in the excludes list, we skip it
      if self.exclude != None and test.name in self.exclude:
        continue
      
      print textwrap.dedent("""
      =====================================================
        Beginning {name}
      -----------------------------------------------------
      """.format(name=test.name))

      ##########################
      # Start this test
      ##########################  
      print textwrap.dedent("""
      -----------------------------------------------------
        Starting {name}
      -----------------------------------------------------
      """.format(name=test.name))
      try:
        p = subprocess.Popen(self.ssh_string, stdin=subprocess.PIPE, shell=True)
        p.communicate("""
          sudo restart mysql
          sudo restart mongodb
        """)
        time.sleep(10)
        
        result = test.start()
        if result != 0: 
          test.stop()
          time.sleep(5)
          print "ERROR: Problem starting " + test.name
          print textwrap.dedent("""
            -----------------------------------------------------
              Stopped {name}
            -----------------------------------------------------
            """.format(name=test.name))
          continue
        
        time.sleep(self.sleep)

        ##########################
        # Verify URLs
        ##########################
        print textwrap.dedent("""
        -----------------------------------------------------
          Verifying URLs for {name}
        -----------------------------------------------------
        """.format(name=test.name))
        test.verify_urls()

        ##########################
        # Benchmark this test
        ##########################
        if self.mode == "benchmark":
          print textwrap.dedent("""
            -----------------------------------------------------
              Benchmarking {name} ...
            -----------------------------------------------------
            """.format(name=test.name))
          test.benchmark()

        ##########################
        # Stop this test
        ##########################
        test.stop()
        time.sleep(5)
        print textwrap.dedent("""
        -----------------------------------------------------
          Stopped {name}
        -----------------------------------------------------
        """.format(name=test.name))
        time.sleep(5)
      except (KeyboardInterrupt, SystemExit):
        test.stop()
        print """
        -----------------------------------------------------
          Cleaning up....
        -----------------------------------------------------
        """
        self.__finish()
        sys.exit()
    
  ############################################################
  # End __run_tests
  ############################################################

  ############################################################
  # __parse_results
  # Ensures that the system has all necessary software to run
  # the tests. This does not include that software for the individual
  # test, but covers software such as curl and weighttp that
  # are needed.
  ############################################################
  def __parse_results(self, tests):
    # Time to create parsed files
    # Aggregate JSON file
    with open(os.path.join(self.full_results_directory(), "results.json"), "w") as f:
      f.write(json.dumps(self.results))
    
    # JSON CSV
    with open(os.path.join(self.full_results_directory(), "json.csv"), 'wb') as csvfile:
      writer = csv.writer(csvfile)
      writer.writerow(["Framework"] + self.concurrency_levels)
      for key, value in self.results['rawData']['json'].iteritems():
        framework = self.results['frameworks'][int(key)]
        writer.writerow([framework] + value)

    # DB CSV
    with open(os.path.join(self.full_results_directory(), "db.csv"), 'wb') as csvfile:
      writer = csv.writer(csvfile)
      writer.writerow(["Framework"] + self.concurrency_levels)
      for key, value in self.results['rawData']['db'].iteritems():
        framework = self.results['frameworks'][int(key)]
        writer.writerow([framework] + value)

    # Query CSV
    with open(os.path.join(self.full_results_directory(), "query.csv"), 'wb') as csvfile:
      writer = csv.writer(csvfile)
      writer.writerow(["Framework"] + self.query_intervals)
      for key, value in self.results['rawData']['query'].iteritems():
        framework = self.results['frameworks'][int(key)]
        writer.writerow([framework] + value)

    # Fortune CSV
    with open(os.path.join(self.full_results_directory(), "fortune.csv"), 'wb') as csvfile:
      writer = csv.writer(csvfile)
      writer.writerow(["Framework"] + self.query_intervals)
      if 'fortune' in self.results['rawData'].keys():
        for key, value in self.results['rawData']['fortune'].iteritems():
          framework = self.results['frameworks'][int(key)]
          writer.writerow([framework] + value)

  ############################################################
  # End __parse_results
  ############################################################

  ############################################################
  # __finish
  ############################################################
  def __finish(self):
    print "Time to complete: " + str(int(time.time() - self.start_time)) + " seconds"
    print "Results are saved in " + os.path.join(self.result_directory, self.timestamp)

  ############################################################
  # End __finish
  ############################################################

  ##########################################################################################
  # Constructor
  ########################################################################################## 

  ############################################################
  # Initialize the benchmarker. The args are the arguments 
  # parsed via argparser.
  ############################################################
  def __init__(self, args):
    self.__dict__.update(args)
    self.start_time = time.time()

    # setup some additional variables
    if self.database_host == None: self.database_host = self.client_host

    self.result_directory = os.path.join("results", self.name)
      
    if self.parse != None:
      self.timestamp = self.parse
    else:
      self.timestamp = time.strftime("%Y%m%d%H%M%S", time.localtime())

    # Setup the concurrency levels array. This array goes from
    # starting_concurrency to max concurrency, doubling each time
    self.concurrency_levels = []
    concurrency = self.starting_concurrency
    while concurrency <= self.max_concurrency:
      self.concurrency_levels.append(concurrency)
      concurrency = concurrency * 2

    # Setup query interval array
    # starts at 1, and goes up to max_queries, using the query_interval
    self.query_intervals = []
    queries = 1
    while queries <= self.max_queries:
      self.query_intervals.append(queries)
      if queries == 1:
        queries = 0

      queries = queries + self.query_interval
    
    # Load the latest data
    self.latest = None
    try:
      with open('latest.json', 'r') as f:
        # Load json file into config object
        self.latest = json.load(f)
    except IOError:
      pass
    
    self.results = None
    try:
      if self.latest != None and self.name in self.latest.keys():
        with open(os.path.join(self.result_directory, str(self.latest[self.name]), 'results.json'), 'r') as f:
          # Load json file into config object
          self.results = json.load(f)
    except IOError:
      pass
    
    if self.results == None:
      self.results = dict()
      self.results['concurrencyLevels'] = self.concurrency_levels
      self.results['queryIntervals'] = self.query_intervals
      self.results['frameworks'] = [t.name for t in self.__gather_tests()]
      self.results['rawData'] = dict()
      self.results['rawData']['json'] = dict()
      self.results['rawData']['db'] = dict()
      self.results['rawData']['query'] = dict()
      self.results['rawData']['fortune'] = dict()
      self.results['weighttpData'] = dict()
      self.results['weighttpData']['json'] = dict()
      self.results['weighttpData']['db'] = dict()
      self.results['weighttpData']['query'] = dict()
      self.results['weighttpData']['fortune'] = dict()
    else:
      for x in self.__gather_tests():
        if x.name not in self.results['frameworks']:
          self.results['frameworks'] = self.results['frameworks'] + [x.name]

    # Setup the ssh command string
    self.ssh_string = "ssh -T -o StrictHostKeyChecking=no " + self.client_user + "@" + self.client_host
    if self.identity_file != None:
      self.ssh_string = self.ssh_string + " -i " + self.identity_file

    if self.install_software:
      install = Installer(self)
      install.install_software()

  ############################################################
  # End __init__
  ############################################################