Basic distributed filesystem write performance test.

dfs-test-write.py

#!/usr/env python
import uuid
import time
import sys
import getopt

def main():
    try:
        opts, args = getopt.getopt(sys.argv[1:], "fd:v", ["files=", "directory="])

    except getopt.GetoptError, err:
        print str(err)
        print usage()
        sys.exit(2)

    args = {}
    for o,a in opts:
        if o == "--files":
            args[o] = int(a)
        elif o == "--directory":
            args[o] = str(a)

    if "--files" not in args or "--directory" not in args:
        print usage()
        sys.exit(2)

    total_start = time.time()
    resultset = []
    failed_writes = 0

    for i in range(args['--files']):
        id = uuid.uuid1()
        filename = "%s/test_%s" % (args['--directory'], id)
        print "writing to %s..." % filename
        testdata = "content-%s" % id

        start_time = time.time()
        try:
            with open(filename, 'w+') as f:
                f.write(testdata)
        except IOError as (errno, strerror):
            failed_writes += 1

        single_write = float(time.time() - start_time)
        print "Took %s seconds" % single_write
        resultset.append(single_write)

    print "\n----------------------------------------"
    print "Files written: %d" % len(resultset)
    print "Median write time: %.9f seconds" % median(resultset)
    print "Fastest write: %.9f seconds" % min(float(n) for n in resultset)
    print "Slowest write: %.9f seconds" % max(float(n) for n in resultset)
    print "Failed writes: %d" % failed_writes
    print "Total execution time: %.3f seconds" % (time.time() - total_start)
    print "----------------------------------------"

def usage(program_name=None):
    return """
Usage: %s [arguments]

        --files <int>                   Number of files to attempt writing.
        --directory <string>            Directory to write test files to.

""" % program_name

def median(pool):
    copy = sorted(pool)
    size = len(copy)
    if size % 2 == 1:
        return copy[(size - 1) / 2]
    else:
        return (copy[size/2 - 1] + copy[size/2]) / 2

if __name__ == "__main__":
    main()