perfkitbenchmarker/linux_benchmarks/bonnie_benchmark.py

# Copyright 2014 PerfKitBenchmarker 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.

"""Runs plain vanilla bonnie++."""


import logging

from perfkitbenchmarker import configs
from perfkitbenchmarker import regex_util
from perfkitbenchmarker import sample


BENCHMARK_NAME = 'bonnieplusplus'
BENCHMARK_CONFIG = """
bonnieplusplus:
  description: >
      Runs Bonnie++. Running this benchmark inside
      a container is currently not supported,
      since Docker tries to run it as root, which
      is not recommended.
  vm_groups:
    default:
      vm_spec: *default_dual_core
      disk_spec: *default_500_gb
"""

LATENCY_REGEX = r'([0-9]*\.?[0-9]+)(\w+)'


# Bonnie++ result fields mapping, see man bon_csv2txt for details.
BONNIE_RESULTS_MAPPING_1_96 = {
    'format_version': 0,
    'bonnie_version': 1,
    'name': 2,
    'concurrency': 3,
    'seed': 4,
    'file_size': 5,
    'chunk_size': 6,
    'putc': 7,
    'putc_cpu': 8,
    'put_block': 9,
    'put_block_cpu': 10,
    'rewrite': 11,
    'rewrite_cpu': 12,
    'getc': 13,
    'getc_cpu': 14,
    'get_block': 15,
    'get_block_cpu': 16,
    'seeks': 17,
    'seeks_cpu': 18,
    'num_files': 19,
    'max_size': 20,
    'min_size': 21,
    'num_dirs': 22,
    'file_chunk_size': 23,
    'seq_create': 24,
    'seq_create_cpu': 25,
    'seq_stat': 26,
    'seq_stat_cpu': 27,
    'seq_del': 28,
    'seq_del_cpu': 29,
    'ran_create': 30,
    'ran_create_cpu': 31,
    'ran_stat': 32,
    'ran_stat_cpu': 33,
    'ran_del': 34,
    'ran_del_cpu': 35,
    'putc_latency': 36,
    'put_block_latency': 37,
    'rewrite_latency': 38,
    'getc_latency': 39,
    'get_block_latency': 40,
    'seeks_latency': 41,
    'seq_create_latency': 42,
    'seq_stat_latency': 43,
    'seq_del_latency': 44,
    'ran_create_latency': 45,
    'ran_stat_latency': 46,
    'ran_del_latency': 47,
}

# Bonnie 1.97 looks the same as 1.96 as far as headings
BONNIE_RESULTS_MAPPING_1_97 = BONNIE_RESULTS_MAPPING_1_96

BONNIE_RESULTS_MAPPING_1_98 = {
    'format_version': 0,
    'bonnie_version': 1,
    'name': 2,
    'concurrency': 3,
    'seed': 4,
    'file_size': 5,
    'chunk_size': 6,
    'seeks_count': 7,
    'seek_proc_count': 8,
    'putc': 9,
    'putc_cpu': 10,
    'put_block': 11,
    'put_block_cpu': 12,
    'rewrite': 13,
    'rewrite_cpu': 14,
    'getc': 15,
    'getc_cpu': 16,
    'get_block': 17,
    'get_block_cpu': 18,
    'seeks': 19,
    'seeks_cpu': 20,
    'num_files': 21,
    'max_size': 22,
    'min_size': 23,
    'num_dirs': 24,
    'file_chunk_size': 25,
    'seq_create': 26,
    'seq_create_cpu': 27,
    'seq_stat': 28,
    'seq_stat_cpu': 29,
    'seq_del': 30,
    'seq_del_cpu': 31,
    'ran_create': 32,
    'ran_create_cpu': 33,
    'ran_stat': 34,
    'ran_stat_cpu': 35,
    'ran_del': 36,
    'ran_del_cpu': 37,
    'putc_latency': 38,
    'put_block_latency': 39,
    'rewrite_latency': 40,
    'getc_latency': 41,
    'get_block_latency': 42,
    'seeks_latency': 43,
    'seq_create_latency': 44,
    'seq_stat_latency': 45,
    'seq_del_latency': 46,
    'ran_create_latency': 47,
    'ran_stat_latency': 48,
    'ran_del_latency': 49,
}

BONNIE_SUPPORTED_VERSIONS = {
    '1.96': BONNIE_RESULTS_MAPPING_1_96,
    '1.97': BONNIE_RESULTS_MAPPING_1_97,
    '1.98': BONNIE_RESULTS_MAPPING_1_98,
}


def GetConfig(user_config):
  return configs.LoadConfig(BENCHMARK_CONFIG, user_config, BENCHMARK_NAME)


def Prepare(benchmark_spec):
  """Install Bonnie++ on the target vm.

  Args:
    benchmark_spec: The benchmark specification. Contains all data that is
      required to run the benchmark.
  """
  vms = benchmark_spec.vms
  vm = vms[0]
  logging.info('Bonnie++ prepare on %s', vm)
  vm.InstallPackages('bonnie++')


def IsValueValid(value):
  """Validate the value.

  An invalid value is either an empty string or a string of multiple '+'.

  Args:
    value: string. The value in raw result.

  Returns:
    A boolean indicates if the value is valid or not.
  """
  if value == '' or '+' in value:
    return False
  return True


def IsCpuField(field):
  """Check if the field is cpu percentage.

  Args:
    field: string. The name of the field.

  Returns:
    A boolean indicates if the field contains keyword 'cpu'.
  """
  return 'cpu' in field


def IsLatencyField(field):
  """Check if the field is latency.

  Args:
    field: string. The name of the field.

  Returns:
    A boolean indicates if the field contains keyword 'latency'.
  """
  return 'latency' in field


def ParseLatencyResult(result):
  """Parse latency result into value and unit.

  Args:
    result: string. Latency value in string format, contains value and unit. eg.
      200ms

  Returns:
    A tuple of value (float) and unit (string).
  """
  match = regex_util.ExtractAllMatches(LATENCY_REGEX, result)[0]
  return float(match[0]), match[1]


def UpdateMetadata(metadata, key, value):
  """Check if the value is valid, update metadata with the key, value pair.

  Args:
    metadata: dict. A dictionary of sample metadata.
    key: string. Key that will be added into metadata dictionary.
    value: Value that of the key.
  """
  if IsValueValid(value):
    metadata[key] = value


def CreateSamples(
    results, start_index, end_index, metadata, field_index_mapping
):
  """Create samples with data in results from start_index to end_index.

  Args:
    results: A list of string representing bonnie++ results.
    start_index: integer. The start index in results list of the samples.
    end_index: integer. The end index in results list of the samples.
    metadata: dict. A dictionary of metadata added into samples.
    field_index_mapping: dict. A dictionary maps field index to field names.

  Returns:
    A list of sample.Sample instances.
  """
  samples = []
  for field_index in range(start_index, end_index):
    field_name = field_index_mapping[field_index]
    value = results[field_index]
    if not IsValueValid(value):
      continue
    if IsCpuField(field_name):
      unit = '%s'
    elif IsLatencyField(field_name):
      value, unit = ParseLatencyResult(value)
    else:
      unit = 'K/sec'
    samples.append(sample.Sample(field_name, float(value), unit, metadata))
  return samples


def ParseCSVResults(results):
  """Parse csv format bonnie++ results.

  Sample Results:
    1.96,1.96,perfkit-7b22f510-0,1,1421800799,7423M,,,,72853,15,47358,5,,,
    156821,7,537.7,10,100,,,,,49223,58,+++++,+++,54405,53,2898,97,+++++,+++,
    59089,60,,512ms,670ms,,44660us,200ms,3747us,1759us,1643us,33518us,192us,
    839us

  Args:
    results: string. Bonnie++ results.

  Returns:
    A list of samples in the form of 3 or 4 tuples. The tuples contain
        the sample metric (string), value (float), and unit (string).
        If a 4th element is included, it is a dictionary of sample
        metadata.
  """
  results = results.split(',')

  format_version = results[0]

  if format_version in BONNIE_SUPPORTED_VERSIONS:
    bonnie_results_mapping = BONNIE_SUPPORTED_VERSIONS[format_version]
    logging.info('Detected bonnie++ CSV format version %s', format_version)
  else:
    raise ValueError(
        f'Unsupported bonnie++ CSV Format version: {format_version} '
        f'(expected version {BONNIE_SUPPORTED_VERSIONS.keys()})'
    )

  field_index_mapping = {}
  for field, value in bonnie_results_mapping.items():
    field_index_mapping[value] = field
  assert len(results) == len(bonnie_results_mapping)
  samples = []
  metadata = {}
  for field_index in range(
      bonnie_results_mapping['format_version'],
      bonnie_results_mapping['chunk_size'] + 1,
  ):
    UpdateMetadata(
        metadata, field_index_mapping[field_index], results[field_index]
    )

  for field_index in range(
      bonnie_results_mapping['num_files'],
      bonnie_results_mapping['file_chunk_size'] + 1,
  ):
    UpdateMetadata(
        metadata, field_index_mapping[field_index], results[field_index]
    )
  samples.extend(
      CreateSamples(
          results,
          bonnie_results_mapping['putc'],
          bonnie_results_mapping['num_files'],
          metadata,
          field_index_mapping,
      )
  )
  samples.extend(
      CreateSamples(
          results,
          bonnie_results_mapping['seq_create'],
          bonnie_results_mapping['ran_del_latency'] + 1,
          metadata,
          field_index_mapping,
      )
  )
  return samples


def Run(benchmark_spec):
  """Run Bonnie++ on the target vm.

  Args:
    benchmark_spec: The benchmark specification. Contains all data that is
      required to run the benchmark.

  Returns:
    A list of samples in the form of 3 or 4 tuples. The tuples contain
        the sample metric (string), value (float), and unit (string).
        If a 4th element is included, it is a dictionary of sample
        metadata.
  """
  vms = benchmark_spec.vms
  vm = vms[0]
  logging.info('Bonnie++ running on %s', vm)
  bonnie_command = '/usr/sbin/bonnie++ -q -d %s -s %d -n 100 -f' % (
      vm.GetScratchDir(),
      2 * vm.total_memory_kb / 1024,
  )
  logging.info('Bonnie++ Results:')
  out, _ = vm.RemoteCommand(bonnie_command)
  return ParseCSVResults(out.strip())


def Cleanup(benchmark_spec):
  """Cleanup Bonnie++ on the target vm (by uninstalling).

  Args:
    benchmark_spec: The benchmark specification. Contains all data that is
      required to run the benchmark.
  """
  pass