Cloud Queues is an open-source messaging system, designed to help customers build and scale distributed applications in the cloud. The Cloud Queues API is built to be flexible, supporting a variety of messaging patterns like producer-consumer, poll based publisher-subscriber, and point-to-point.
Queuing systems like Cloud Queues are meant to decouple resources in an application by serving as a communication layer between those resources. Instead of a server in an application telling another server do to a task, they post a message to a queue. Then, the server performing work can check that queue for work when it is available. This prevents an application from slowing down due to servers being busy, offline, unavailable, etc. Using Cloud Queues as a middle layer means that servers producing work can post messages to the queue as quickly as needed, without caring if a server is available to do that work. Servers working on tasks can easily be scaled up and down as needed.
Most distributed applications will have the need for a queuing system, but here are some specific examples for some common messaging patterns:
Producer-Consumer- A producer-consumer model is the most popular way to use a queuing system. A good example of this is a website that allows users to signup for the company's newsletter. As customers fill out the online form, they need to be sent a confirmation email to ensure they own the email address. Web servers responsible for running the website will place a message on a queue every time a customer submits their form, "send_conf_email" for example. Applications servers responsible for sending email will be checking the queue every minute to see if any new jobs are ready to for work. A server will claim the message, ensuring no other servers attempt to do the same work, it will complete the tasks, and then delete the message, signifying the task is complete. Many applications will use multiple queues for different types of tasks. For example, another job called "update_salesforce" may be triggered by this same event.
Publisher-Subscriber- The publisher-subscriber pattern is often used as the underlying component of a notifications service. One of the most common examples is an RSS feed. Customers signup online to receive alerts when an event happens, for example, when their company name is mentioned in an online article. In this example, there is an agent looking online at all times for certain keywords. As it finds articles, a message will be posted to the queue that contains the article's information. Servers responsible for alerting, the "subscribers" in this case, will poll the queue every 1-5 minutes to see if any new updates have been posted. If so, they will email anyone who has signed up for updates.
Please see the Cloud Queues FAQ for more information on features, limits, support, and more.
© 2011-2013 Rackspace US, Inc.
Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License