Created
October 24, 2014 21:54
-
-
Save ndimiduk/91af1dbc4f7ca815f21d to your computer and use it in GitHub Desktop.
Parse the log output produced by IntegrationTestRegionReplicaPerf for latency summary data
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
# | |
# parse the log output produced by IntegrationTestRegionReplicaPerf to | |
# collect the latency data from each client thread. Works on some build of | |
# 0.98 + region replicas. | |
# | |
import fileinput | |
import re | |
NEW_RUN = re.compile('.*Launching.*') | |
IS_NONREPLICA = re.compile('.* non-replica .*') | |
JOB_NUM = re.compile('.*(\d)/3$') | |
TEST_CLIENT = re.compile('.*TestClient-(\d).*') | |
FOUR_9S = re.compile('.*99\.99% = (.*)$') | |
THREE_9S = re.compile('.*99\.9% = (.*)$') | |
TWO_9S = re.compile('.*99% = (.*)$') | |
EIGHTY = re.compile('.*80% = (.*)$') | |
STDEV = re.compile('.*Standard Deviation = (.*)$') | |
MEAN = re.compile('.*Mean = (.*)$') | |
cur_job = None | |
isReplica = None | |
clients = None | |
for line in fileinput.input(): | |
line = line.strip() | |
if not (NEW_RUN.match(line) or FOUR_9S.match(line) or THREE_9S.match(line) or TWO_9S.match(line) or EIGHTY.match(line) or STDEV.match(line) or MEAN.match(line)): continue | |
if NEW_RUN.match(line): | |
found_first_run = True | |
if IS_NONREPLICA.match(line): | |
isReplica = False | |
else: | |
isReplica = True | |
cur_job = JOB_NUM.match(line).group(1) | |
if clients: | |
for (thread,run) in clients.iteritems(): | |
print "{} {} {} {} {} {} {} {} {}".format(run['replica'], run['run'], run['client'], run['stdev'], run['mean'], run['80pct'], run['99pct'], run['99.9pct'], run['99.99pct']) | |
clients = {} | |
if TEST_CLIENT.match(line): | |
c = TEST_CLIENT.match(line).group(1) | |
x = clients.get(c, None) | |
if not x: | |
clients[c] = {} | |
clients[c]['client'] = c | |
clients[c]['replica'] = isReplica | |
clients[c]['run'] = cur_job | |
if FOUR_9S.match(line): | |
clients[c]['99.99pct'] = FOUR_9S.match(line).group(1) | |
continue | |
if THREE_9S.match(line): | |
clients[c]['99.9pct'] = THREE_9S.match(line).group(1) | |
continue | |
if TWO_9S.match(line): | |
clients[c]['99pct'] = TWO_9S.match(line).group(1) | |
continue | |
if EIGHTY.match(line): | |
clients[c]['80pct'] = EIGHTY.match(line).group(1) | |
continue | |
if MEAN.match(line): | |
clients[c]['mean'] = MEAN.match(line).group(1) | |
continue | |
if STDEV.match(line): | |
clients[c]['stdev'] = STDEV.match(line).group(1) | |
continue | |
if clients: | |
for (thread,run) in clients.iteritems(): | |
print "{} {} {} {} {} {} {} {} {}".format(run['replica'], run['run'], run['client'], run['stdev'], run['mean'], run['80pct'], run['99pct'], run['99.9pct'], run['99.99pct']) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment