zerofun 1.0.0

Remote function calls for array data using ZMQ

zerofun

Remote function calls for array data using [ZMQ](Remote function calls for array data using ZMQ).

Overview

Zerofun provides a Server that you can bind functions to and a Client that can call the messages and receive their results. The function inputs and results are both flat dicts of Numpy arrays. The data is sent efficiently without serialization to maximize throughput.

Installation

pip install zerofun

Example

Server:

import zerofun

def add(data):
  return {'result': data['foo'] + data['bar']}

def msg(data):
  print('Message from client:', data['msg'])

server = zerofun.Server('tcp://*:2222', workers=1, ipv6=False)
server.bind('add', add)
server.bind('msg', msg)
server.run()

Client:

import zerofun

client = zerofun.Client('tcp://localhost:2222', ipv6=False, maxinflight=16)
client.connect()

future = client.add({'foo': 1, 'bar': 1})
result = future.result()
print(result)  # {'result': 2}

client.msg()

API

Client(address, identity=None, name='Client', ipv6=False, pings=10,
       maxage=120, maxinflight=16, errors=True, connect=False)
  connect(retry=True, timeout=10)  # Open connection to a server.
  call(method, data)  # Call remote function by name and input data.
  __getattr__(name)   # Syntactic sugar to call remote functions.
  close()             # Close the connection.
  stats()             # Return a dict of client statistics.

Server(address, workers=1, name='Server', errors=True, ipv6=False)
  bind(name, workfn, donefn=None, workers=0, batch=0)  # Register a function.
  start()  # Start the server.
  check()  # Check for exceptions in server functions and reraise them.
  close()  # Stop the server.
  run()    # Start the server and keep checking for exceptions forever.
  stats()  # Return a dict of server statistics.

Features

Several productivity and performance features are available:

• Request batching: The server can batch requests together so that the user function receives a dict of stacked arrays and the function result will be split and sent back to the corresponding clients.
• Multithreading: Servers can use a thread pool to process multiple requests in parallel. Optionally, each function can also request its own thread pool to allow functions to block (e.g. for rate limiting) without blocking other functions.
• Async requests: Clients can send multiple overlapping requests and wait on the results when needed using Future objects. The maximum number of inflight requests can be limited to avoid requests building up when the server is slower than the client.
• Error handling: Exceptions raised in server functions are reported to the client and raised in future.result() or, if the user did not store the future object, on the next request.
• Heartbeating: Clients can send ping requests when they have not received a result from the server for a while, allowing them to distinguish between a server that is computing a heavy result and a connection loss.
• GIL load reduction: The ProcServer behaves just like the normal Server but uses a background process to batch requests and fan out results, substantially reducing GIL load for the server workers in the main process.

Questions

Please open a GitHub issue for each question. Over time, we will add common questions to the README.