We stared our discussion from the need to perform the tasks in the background and from there, we introduced the queue paradigm of processing tasks that store messages that are processed by worker nodes. Then we focused our attention on the queue model as implemented by Google App Engine. Next, we took a very detailed look into how to configure queues using queue.yaml file. We learned that we can have multiple queues with different task processing rates and retry parameters. Next, we focused on how to add tasks to the queue and how to add information using headers, parameters, and payload. Finally, we did put all this into practice by a URL downloader application that downloads the HTML pages and compresses them.

Once comfortable with push queues, we learned about a helpful library deferred that can remove lots of boilerplate code. So we wrote our earlier application using the aforementioned library and found out that newer version of our code became much simpler.

Finally, we turned...