Building small- to medium-sized distributed applications in Python, as we saw, is not particularly difficult. Once a distributed system grows to a larger size, the design and development effort needed tends to grow as well in a super-linear fashion.

In these cases, a more solid foundation on the theory of distributed systems becomes necessary. There are a number of resources available both online and offline. Most big universities give courses on this subject, and a number of them are freely available online.

One good example is the ETH course on Principles of Distributed Computing (http://dcg.ethz.ch/lectures/podc_allstars/index.html), which covers a number of fundamentals, including synchronization, consensus, and eventual consistency (including the famous CAP theorem).

Having said that, beginners should not feel discouraged. The gain in performance that even a few lines of code in a simple framework, such as Python-RQ, can give to our code is just astounding!