--- /dev/null
+#!/usr/bin/env python
+
+import matplotlib
+import pylab as plt
+import numpy as np
+
+import os
+import sys
+import math
+
+if __name__ == '__main__':
+ files = sys.argv[1:]
+ for f in files:
+ execfile(f)
+
+ #names = ['scale', 'multipart:pct', 'multipart:cpu']
+
+ def deal_with_posK_res(k, x):
+ if type(x) == list:
+ return [e[k] for e in x]
+ return x[k]
+
+ def deal_with_pos0_res(x):
+ return deal_with_posK_res(0, x)
+
+ def deal_with_pos1_res(x):
+ return deal_with_posK_res(1, x)
+
+ import re
+ RGX = re.compile(r'--new-order-remote-item-pct (\d+)')
+ def extract_pct(x):
+ m = RGX.search(x)
+ assert m
+ p = int(m.group(1))
+ assert p >= 0 and p <= 100
+ def pn(n, p):
+ return 1.0 - (1.0 - p)**n
+ def ex(p):
+ import math
+ return math.fsum([(1.0/11.0)*pn(float(n), p) for n in range(5, 16)])
+ return ex(p/100.0) * 100.0
+
+ def extract_p(x):
+ m = RGX.search(x)
+ assert m
+ p = int(m.group(1))
+ assert p >= 0 and p <= 100
+ return p
+
+ def multipart_cpu_process(config):
+ assert config['db'] == 'ndb-proto2' or \
+ config['db'] == 'kvdb'
+ if config['db'] == 'ndb-proto2':
+ return config['threads']
+ else:
+ return 8
+
+ def readonly_lines_func(config):
+ if 'disable-read-only-snapshots' in config['bench_opts']:
+ return 'No-Snapshots'
+ else:
+ return 'Snapshots'
+
+ def MFormatter(x, p):
+ if x == 0:
+ return '0'
+ v = float(x)/float(10**6)
+ if math.ceil(v) == v:
+ return '%dM' % v
+ return '%.1fM' % v
+
+ def KFormatter(x, p):
+ if x == 0:
+ return '0'
+ v = float(x)/float(10**3)
+ if math.ceil(v) == v:
+ return '%dK' % v
+ return '%.1fK' % v
+
+ descs = [
+ {
+ 'name' : 'scale',
+ 'x-axis' : 'threads',
+ 'x-axis-func' : lambda x: x,
+ 'y-axis' : deal_with_pos0_res,
+ 'lines' : ['db'], # each line holds this constant
+ 'x-label' : 'threads',
+ 'y-label' : 'throughput (txns/sec)',
+ 'y-axis-major-formatter' : matplotlib.ticker.FuncFormatter(MFormatter),
+ 'x-axis-set-major-locator' : True,
+ #'title' : 'ycsb throughput graph',
+ },
+ {
+ 'name' : 'scale_tpcc',
+ 'x-axis' : 'threads',
+ 'x-axis-func' : lambda x: x,
+ 'y-axis' : deal_with_pos0_res,
+ 'lines' : ['db'], # each line holds this constant
+ 'x-label' : 'threads',
+ 'y-label' : 'throughput (txns/sec)',
+ 'y-axis-major-formatter' : matplotlib.ticker.FuncFormatter(MFormatter),
+ 'x-axis-set-major-locator' : True,
+ #'title' : 'tpcc throughput graph',
+ },
+ {
+ 'name' : 'multipart:pct',
+ 'x-axis' : 'bench_opts',
+ 'x-axis-func' : extract_pct,
+ 'y-axis' : deal_with_pos0_res,
+ 'lines' : ['db'], # each line holds this constant
+ 'x-label' : '% cross-partition',
+ 'y-label' : 'throughput (txns/sec)',
+ 'y-axis-major-formatter' : matplotlib.ticker.FuncFormatter(MFormatter),
+ 'x-axis-set-major-locator' : False,
+ #'title' : 'tpcc new-order throughput graph',
+ 'legend' : 'upper right',
+ },
+ #{
+ # 'name' : 'multipart:cpu',
+ # 'x-axis-process' : multipart_cpu_process,
+ # 'y-axis' : deal_with_pos0_res,
+ # 'lines' : ['db'], # each line holds this constant
+ # 'x-label' : 'num threads',
+ # 'y-label' : 'txns/sec',
+ # 'title' : 'tpcc full workload throughput graph',
+ #},
+ {
+ 'name' : 'readonly',
+ 'x-axis' : 'bench_opts',
+ 'x-axis-func' : extract_p,
+ 'y-axis' : deal_with_pos0_res,
+ 'lines-func' : readonly_lines_func,
+ 'x-label' : '% remote warehouse stock',
+ 'y-label' : 'throughput (txns/sec)',
+ 'y-axis-major-formatter' : matplotlib.ticker.FuncFormatter(KFormatter),
+ 'x-axis-set-major-locator' : True,
+ #'title' : 'tpcc read only throughput graph',
+ 'legend' : 'right',
+ },
+ ]
+
+ def label_transform(x):
+ if x == 'kvdb':
+ return 'Key-Value'
+ if x == 'ndb-proto1':
+ return 'Malflingo-Star'
+ if x == 'ndb-proto2':
+ return 'Malflingo'
+ if x == 'kvdb-st':
+ return 'Partitioned-Store'
+ return x
+
+ for desc in descs:
+ bench = desc['name']
+ bench_results = [d for d in RESULTS if d[0]['name'] == bench]
+ if not bench_results:
+ print >>sys.stderr, 'skipping bench %s' % bench
+ continue
+ lines = {}
+ for (config, result) in bench_results:
+ if 'lines-func' in desc:
+ key = desc['lines-func'](config)
+ else:
+ key = tuple(config[x] for x in desc['lines'])
+ pts = lines.get(key, {})
+ if 'x-axis-process' in desc:
+ xpt = desc['x-axis-process'](config)
+ else:
+ xpt = desc['x-axis-func'](config[desc['x-axis']])
+ assert xpt not in pts
+ pts[xpt] = desc['y-axis'](result)
+ lines[key] = pts
+
+ def mean(x): return sum(x)/len(x)
+ def median(x): return sorted(x)[len(x)/2]
+
+ # find min/max of xpts
+ xmin = min([e for l in lines.values() for e in l])
+ xmax = max([e for l in lines.values() for e in l])
+ #print xmin, xmax
+
+ labels = []
+ for (name, pts) in lines.iteritems():
+ spts = sorted(pts.iteritems(), key=lambda x: x[0])
+ ypts = [sorted(x[1]) for x in spts]
+ ymins = np.array([min(x) for x in ypts])
+ ymaxs = np.array([max(x) for x in ypts])
+ ymid = np.array([median(x) for x in ypts])
+ yerr=np.array([ymid - ymins, ymaxs - ymid])
+ xpts = [x[0] for x in spts]
+ assert len(xpts)
+ if len(xpts) == 1:
+ xpts = range(xmin, xmax + 1)
+ assert len(ymins) == 1
+ assert len(ymaxs) == 1
+ assert len(ymid) == 1
+ ymins = np.array([ymins[0] for _ in xpts])
+ ymaxs = np.array([ymaxs[0] for _ in xpts])
+ ymid = np.array([ymid[0] for _ in xpts])
+ yerr=np.array([ymid - ymins, ymaxs - ymid])
+
+ plt.errorbar(xpts, ymid, yerr=yerr)
+ if type(name) == str:
+ labels.append(label_transform(name))
+ else:
+ labels.append(label_transform('-'.join(name)))
+
+ ax = plt.gca()
+ if desc['x-axis-set-major-locator']:
+ ax.xaxis.set_major_locator(matplotlib.ticker.FixedLocator(sorted(lines.values()[0].keys())))
+ if 'y-axis-major-formatter' in desc:
+ ax.yaxis.set_major_formatter(desc['y-axis-major-formatter'])
+
+ plt.xlabel(desc['x-label'])
+ plt.ylabel(desc['y-label'])
+ if 'title' in desc:
+ plt.title(desc['title'])
+
+ plt.xlim(xmin = xmin, xmax = xmax)
+ plt.ylim(ymin = 0)
+
+ placement = 'upper left' if not 'legend' in desc else desc['legend']
+ plt.legend(labels, loc=placement)
+ bname = '.'.join(os.path.basename(f).split('.')[:-1])
+ plt.savefig('.'.join([bname + '-' + bench, 'pdf']))
+ plt.close()
projects / c11concurrency-benchmarks.git / blobdiff