常见问题

Why isn’t this example with time.sleep() running in parallel?

Many people’s first foray into Tornado’s concurrency looks something like this:

class BadExampleHandler(RequestHandler):
    def get(self):
        for i in range(5):
            print(i)
            time.sleep(1)

Fetch this handler twice at the same time and you’ll see that the second five-second countdown doesn’t start until the first one has completely finished. The reason for this is that time.sleep is a blocking function: it doesn’t allow control to return to the IOLoop so that other handlers can be run.

Of course, time.sleep is really just a placeholder in these examples, the point is to show what happens when something in a handler gets slow. No matter what the real code is doing, to achieve concurrency blocking code must be replaced with non-blocking equivalents. This means one of three things:

  1. Find a coroutine-friendly equivalent. For time.sleep, use tornado.gen.sleep instead:

    class CoroutineSleepHandler(RequestHandler):
    @gen.coroutine
    def get(self):
        for i in range(5):
            print(i)
            yield gen.sleep(1)

    When this option is available, it is usually the best approach. See the Tornado wiki for links to asynchronous libraries that may be useful.

  2. Find a callback-based equivalent. Similar to the first option, callback-based libraries are available for many tasks, although they are slightly more complicated to use than a library designed for coroutines. These are typically used with tornado.gen.Task as an adapter:

    class CoroutineTimeoutHandler(RequestHandler):
    @gen.coroutine
    def get(self):
        io_loop = IOLoop.current()
        for i in range(5):
            print(i)
            yield gen.Task(io_loop.add_timeout, io_loop.time() + 1)

    Again, the Tornado wiki can be useful to find suitable libraries.

  3. Run the blocking code on another thread. When asynchronous libraries are not available, concurrent.futures.ThreadPoolExecutor can be used to run any blocking code on another thread. This is a universal solution that can be used for any blocking function whether an asynchronous counterpart exists or not:

    executor = concurrent.futures.ThreadPoolExecutor(8)

class ThreadPoolHandler(RequestHandler):
    @gen.coroutine
    def get(self):
        for i in range(5):
            print(i)
            yield executor.submit(time.sleep, 1)

See the Asynchronous I/O chapter of the Tornado user’s guide for more on blocking and asynchronous functions.

My code is asynchronous, but it’s not running in parallel in two browser tabs.

Even when a handler is asynchronous and non-blocking, it can be surprisingly tricky to verify this. Browsers will recognize that you are trying to load the same page in two different tabs and delay the second request until the first has finished. To work around this and see that the server is in fact working in parallel, do one of two things:

  • Add something to your urls to make them unique. Instead of http://localhost:8888 in both tabs, load http://localhost:8888/?x=1 in one and http://localhost:8888/?x=2 in the other.
  • Use two different browsers. For example, Firefox will be able to load a url even while that same url is being loaded in a Chrome tab.