if not sample_set_name in guesstimator.list_sample_sets():
guesstimator.create_sample_set(name=sample_set_name)
while True:
# Record that an operation has occurred.
guesstimator.record(sample_set_name)
# our operation happens to be an increment in Redis, it could be anything.
# this is a convenient operation because it lets us keep track of the
# true number of total operations across multiple workers.
operations = redis.incr("guesstimator_example_actual_operation_count")
# reader.py must have reset the operation count.
if last_operations_count > operations:
guesstimator.writes_performed = 0
last_operations_count = operations
if (operations % 10000) == 0:
# Read the timestamp since metrics started being tracked.
# and the estimated number of samples taken.
timestamp, estimated_operations = guesstimator.read(sample_set_name)
# Using the operation count and the frequency that we record metrics we can guestimate how
# many workers are currently in the system.
estimated_worker_count = int((float(operations) / (float(guesstimator.writes_performed) / 0.05)) + 0.5)
print "Metrics:\n\toperation count: \t\033[32m%s\033[m\n\testimated count:\t\033[31m%s\033[m\n\n\tredis writes performed:\t\t%s\n\testimated worker count: \t%s\n---------" % (
operations,
estimated_operations,
