x/turbo-hipster
Code Issues Proposed changes

Regenerate historical data, add auto generated data config

Browse Source 
Change-Id: I04a2d7117338dad49e84b7d7d91c81ed405e6674
This commit is contained in:
Michael Still 2014-01-01 18:05:37 +11:00
parent 81ebd10bdf
commit 690db7e166
14 changed files with 1265 additions and 976 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
datasets/datasets_devstack_131007/131007_devstack_export/config.json Normal file
View File

@ -0,0 +1,12 @@
{
"database": "nova_dataset_131007_devstack",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova.sql",
"type": "mysql"
}

12
datasets/datasets_devstack_131007/131007_devstack_export/input.json Normal file
View File

@ -0,0 +1,12 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "nova_dataset_131007_devstack",
"seed_data": "nova.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
}
}

12
datasets/datasets_devstack_150/datasets_devstack_150/config.json Normal file
View File

@ -0,0 +1,12 @@
{
"database": "datasets_devstack_150",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova.sql",
"type": "mysql"
}

13
datasets/datasets_devstack_150/datasets_devstack_150/input.json Normal file
View File

@ -0,0 +1,13 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "datasets_devstack_150",
"seed_data": "nova.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
}
}

13
datasets/datasets_trivial_500/nova_trivial_500/config.json Normal file
View File

@ -0,0 +1,13 @@
{
"database": "nova_dataset_trivial_500",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"138": 42.0,
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova_trivial_500.sql",
"type": "mysql"
}

13
datasets/datasets_trivial_500/nova_trivial_500/input.json Normal file
View File

@ -0,0 +1,13 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "nova_dataset_trivial_500",
"seed_data": "nova_trivial_500.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
}
}

13
datasets/datasets_trivial_6000/nova_trivial_6000/config.json Normal file
View File

@ -0,0 +1,13 @@
{
"database": "nova_dataset_trivial_6000",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"152": 74.0,
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova_trivial_6000.sql",
"type": "mysql"
}

13
datasets/datasets_trivial_6000/nova_trivial_6000/input.json Normal file
View File

@ -0,0 +1,13 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "nova_dataset_trivial_6000",
"seed_data": "nova_trivial_6000.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
}
}

30
datasets/datasets_user_001/user_001/config.json Normal file
View File

@ -0,0 +1,30 @@
{
"database": "nova_datasets_user_001",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"135": 62.0,
"138": 44.0,
"149": 87.0,
"152": 241.0,
"159": 86.0,
"205": 51.0,
"206": 63.0,
"216": 102.0,
"230": 53.0,
"_138_bugs": [
1263835
],
"_149_bugs": [
1263836
],
"_205_bugs": [
1263868
],
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova_user_001.sql",
"type": "mysql"
}

23
datasets/datasets_user_001/user_001/input.json Normal file
View File

@ -0,0 +1,23 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "nova_datasets_user_001",
"seed_data": "nova_user_001.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30,
"135": 120,
"138": 180,
"_138_bugs": [1263835],
"149": 240,
"_149_bugs": [1263836],
"152": 300,
"159": 120,
"205": 120,
"_205_bugs": [1263868],
"216": 180
}
}

12
datasets/datasets_user_002/user_002/config.json Normal file
View File

@ -0,0 +1,12 @@
{
"database": "nova_dataset_user_002",
"db_pass": "tester",
"db_user": "nova",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
},
"project": "openstack/nova",
"seed_data": "nova_user_002.sql",
"type": "mysql"
}

13
datasets/datasets_user_002/user_002/input.json Normal file
View File

@ -0,0 +1,13 @@
{
"project": "openstack/nova",
"type": "mysql",
"db_user": "nova",
"db_pass": "tester",
"database": "nova_dataset_user_002",
"seed_data": "nova_user_002.sql",
"logging_conf": "logging.conf",
"maximum_migration_times": {
"default": 30
}
}

1949
results.json
View File

File diff suppressed because it is too large Load Diff

113
turbo_hipster/report_historical.py
View File

@ -16,61 +16,118 @@
import json
import math
import numpy
import os
import sys
def main():
for dataset in ['devstack_131007', 'devstack_150', 'trivial_500',
'trivial_6000', 'user_001', 'user_002']:
process_dataset(dataset)
def process_dataset(dataset):
with open('results.json') as f:
results = json.loads(f.read())
migrations = {}
all_times = {}
for engine in ['mysql', 'percona']:
print
print 'Dataset: %s' % dataset
print 'Engine: %s' % engine
print
for migration in sorted(results[engine]['user_001']):
for migration in sorted(results[engine][dataset]):
times = []
for time in results[engine]['user_001'][migration]:
for i in range(results[engine]['user_001'][migration][time]):
all_times.setdefault(migration, [])
for time in results[engine][dataset][migration]:
for i in range(results[engine][dataset][migration][time]):
times.append(int(time))
all_times[migration].append(int(time))
times = sorted(times)
emit_summary(engine, times, migrations, migration)
np_times = numpy.array(times)
mean = np_times.mean()
stddev = np_times.std()
failed_threshold = int(max(30.0, mean + stddev * 2))
failed = 0
for time in times:
if time > failed_threshold:
failed += 1
migrations.setdefault(migration, {})
migrations[migration][engine] = ('%.02f;%0.2f;%.02f'
% (mean - 2 * stddev,
mean,
mean + 2 * stddev))
if failed_threshold != 30 or failed > 0:
print ('%s: Values range from %s to %s seconds. %d values. '
'Mean is %.02f, stddev is %.02f.\n '
'Recommend max of %d. With this value %.02f%% of tests '
'would have failed.'
% (migration, np_times.min(), np_times.max(),
len(times), mean, stddev, failed_threshold,
failed * 100.0 / len(times)))
print
print 'Dataset: %s' % dataset
print 'Engine: combined'
print
for migration in sorted(all_times.keys()):
emit_summary('combined', all_times[migration], migrations, migration)
with open('results.txt', 'w') as f:
f.write('Migration,mysql,percona\n')
for migration in sorted(migrations.keys()):
f.write('%s' % migration)
for engine in ['mysql', 'percona']:
f.write(',%s' % migrations[migration][engine])
f.write(',%s' % migrations[migration].get(engine, ''))
f.write('\n')
# Write out the dataset config as a json blob
config_path = os.path.join('datasets',
'datasets_%s' % dataset,
omg_hard_to_predict_names(dataset))
with open(os.path.join(config_path, 'input.json')) as f:
config = json.loads(f.read())
for migration in sorted(all_times.keys()):
minimum, mean, maximum, stddev = analyse(all_times[migration])
recommend = mean + 2 * stddev
if recommend > 30.0:
config['maximum_migration_times'][migration] = math.ceil(recommend)
with open(os.path.join(config_path, 'config.json'), 'w') as f:
f.write(json.dumps(config, indent=4, sort_keys=True))
def omg_hard_to_predict_names(dataset):
if dataset.startswith('trivial'):
return 'nova_%s' % dataset
if dataset == 'devstack_150':
return 'datasets_devstack_150'
if dataset == 'devstack_131007':
return '131007_devstack_export'
return dataset
def analyse(times):
np_times = numpy.array(times)
minimum = np_times.min()
mean = np_times.mean()
maximum = np_times.max()
stddev = np_times.std()
return minimum, mean, maximum, stddev
def emit_summary(engine, times, migrations, migration):
minimum, mean, maximum, stddev = analyse(times)
failed_threshold = int(max(30.0, mean + stddev * 2))
failed = 0
for time in times:
if time > failed_threshold:
failed += 1
migrations.setdefault(migration, {})
migrations[migration][engine] = ('%.02f;%0.2f;%.02f'
% (mean - 2 * stddev,
mean,
mean + 2 * stddev))
if failed_threshold != 30 or failed > 0:
print ('%s: Values range from %s to %s seconds. %d values. '
'Mean is %.02f, stddev is %.02f.\n '
'Recommend max of %d. With this value %.02f%% of tests '
'would have failed.'
% (migration, minimum, maximum,
len(times), mean, stddev, failed_threshold,
failed * 100.0 / len(times)))
if __name__ == '__main__':
sys.path.insert(0, os.path.abspath(
os.path.join(os.path.dirname(__file__), '../')))