Designing for Push

When designing for push, keep scaling in mind. Normal computing doesn’t need to scale. When coding is done, an app runs on a device using the local CPU. Should a developer deploy an extra 10,000 copies, there’s no further investment involved other than increased technical support.

Push computing does scale. Whether you have 10,000 or 100,000 or 1,000,000 users matters. That’s because developers must provide the service layer that handles the operations for every unit sold. The more users supported, the greater the costs will be. Consider that these services need to be completely reliable and that consumers will not be tolerant of extended downtimes.

Consider an application with just 10,000 users. It might service a million uses per day, assuming update checks every 15 minutes. More time-critical uses might demand checks every few minutes or even several times a minute. As the computational burden builds, so do the hosting costs. While cloud computing provides an excellent match to these kinds of needs, that kind of solution comes with a real price in development, maintenance, and day-to-day operations.