Вопрос

I need to make asynchronous XML-RPC (client call) to a Cobbler server in Tornado, for example something like this:

cobbler_connection = xmlrpclib.Server("http://my_cobbler_server")

...
profile = yield gen.Task(cobbler_connection.get_profile, profile_name)
...

Right now I am using xmlrpclib which does not seem to make async client calls so it ends up blocking Tornado (or so I hear). I'm a newb when it comes to Tornado and asynchronous programming in general (but please don't give me any links to some instructions or beginner's guide). I've tried googling but I can't get a clear sense of what changes I need to make to xmlrpclib so that I can make these calls asynchronous. Any pointers will be greatly appreciated. Thanks in advance.

Это было полезно?

Решение

It's usually a good idea to check if Twisted has support for various operations, since it can be combined with Tornado using tornado.platform.twisted. For XML-RPC there is following code: http://twistedmatrix.com/documents/13.0.0/web/howto/xmlrpc.html A simple example:

import tornado.httpserver
import tornado.ioloop
import tornado.web
import tornado.platform.twisted

tornado.platform.twisted.install()

from twisted.web.xmlrpc import Proxy
from twisted.internet import reactor

proxy = Proxy('http://advogato.org/XMLRPC')

from tornado.options import define, options
define("port", default=8000, help="run on the given port", type=int)

class IndexHandler(tornado.web.RequestHandler):
    def printValue(self, value):
        self.write(repr(value))
        self.finish()

    def printError(self, error):
        self.write('error: %s' % error) 
        self.finish()

    @tornado.web.asynchronous
    def get(self):
        proxy.callRemote('test.sumprod', 3, 5).addCallbacks(self.printValue, self.printError)

if __name__ == "__main__":
    tornado.options.parse_command_line()
    app = tornado.web.Application(handlers=[(r"/", IndexHandler)])
    http_server = tornado.httpserver.HTTPServer(app)
    http_server.listen(options.port)
    tornado.ioloop.IOLoop.instance().start()

Benchmarking:

$ ab -n 1000 -c 5 http://localhost:8000/
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking localhost (be patient)
Completed 100 requests
Completed 200 requests
Completed 300 requests
Completed 400 requests
Completed 500 requests
Completed 600 requests
Completed 700 requests
Completed 800 requests
Completed 900 requests
Completed 1000 requests
Finished 1000 requests


Server Software:        TornadoServer/3.0.1
Server Hostname:        localhost
Server Port:            8000

Document Path:          /
Document Length:        7 bytes

Concurrency Level:      5
Time taken for tests:   76.529 seconds
Complete requests:      1000
Failed requests:        0
Write errors:           0
Total transferred:      201000 bytes
HTML transferred:       7000 bytes
Requests per second:    13.07 [#/sec] (mean)
Time per request:       382.646 [ms] (mean)
Time per request:       76.529 [ms] (mean, across all concurrent requests)
Transfer rate:          2.56 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.0      0       0
Processing:   302  382 498.9    308    5318
Waiting:      302  382 498.9    308    5318
Total:        302  382 498.9    308    5318

Percentage of the requests served within a certain time (ms)
  50%    308
  66%    310
  75%    311
  80%    313
  90%    320
  95%    334
  98%   1310
  99%   3309
 100%   5318 (longest request)

Straight use of xmlrpclib for comparison:

import tornado.httpserver
import tornado.ioloop
import tornado.web

import xmlrpclib    

server = xmlrpclib.ServerProxy('http://advogato.org/XMLRPC')

from tornado.options import define, options
define("port", default=8000, help="run on the given port", type=int)

class IndexHandler(tornado.web.RequestHandler):
    def get(self):
        self.write(repr(server.test.sumprod(3, 5)))

if __name__ == "__main__":
    tornado.options.parse_command_line()
    app = tornado.web.Application(handlers=[(r"/", IndexHandler)])
    http_server = tornado.httpserver.HTTPServer(app)
    http_server.listen(options.port)
    tornado.ioloop.IOLoop.instance().start()

Benchmarking:

$  ab -n 1000 -c 5 http://localhost:8000/
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking localhost (be patient)
Completed 100 requests
Completed 200 requests
Completed 300 requests
Completed 400 requests
Completed 500 requests
Completed 600 requests
Completed 700 requests
Completed 800 requests
Completed 900 requests
Completed 1000 requests
Finished 1000 requests


Server Software:        TornadoServer/3.0.1
Server Hostname:        localhost
Server Port:            8000

Document Path:          /
Document Length:        7 bytes

Concurrency Level:      5
Time taken for tests:   325.538 seconds
Complete requests:      1000
Failed requests:        0
Write errors:           0
Total transferred:      201000 bytes
HTML transferred:       7000 bytes
Requests per second:    3.07 [#/sec] (mean)
Time per request:       1627.690 [ms] (mean)
Time per request:       325.538 [ms] (mean, across all concurrent requests)
Transfer rate:          0.60 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.0      0       0
Processing:   305 1625 484.5   1532    4537
Waiting:      305 1624 484.5   1532    4537
Total:        305 1625 484.5   1532    4537

Percentage of the requests served within a certain time (ms)
  50%   1532
  66%   1535
  75%   1537
  80%   1539
  90%   1547
  95%   1984
  98%   4524
  99%   4533
 100%   4537 (longest request)

Much worse.

Лицензировано под: CC-BY-SA с атрибуция
Не связан с StackOverflow
scroll top