I am hesitant to post this because design patterns are akin to religion for most who read this group so it is often best to keep opinions to one's self. However, since this goes way back to a very old Guido blog post, and because it might be helpful I will take the risk.
Undoubtedly many saw this article about high-levels of function calls affecting Chrome's performance: http://aptiverse.com/blog/closer_look_at_chrome/ A long time ago, Guido posted on his blog a brief study he did showing that high-levels of function calls can really affect Python performance due to overhead costs. I wish I had bookmarked it, but failed to do so. However, it is very easy to setup and test yourself using examples similar to what Guido did. A pernicious design pattern he showed being affected by this is a function call or lambda embedded inside an iterative loop. As I noted, I am hesitant to say this, but given the relatively modest hardware specs that many F1 instance GAE developers deal with, one should very likely test for performance when one's code is performing very high levels for function calls. Certainly if you have code averaging "2.1 effective" lines of Python per function (as noted in the article), and you are running on a F1 with decent QPS, then give it some thought. -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to google-appengine+unsubscr...@googlegroups.com. To post to this group, send email to google-appengine@googlegroups.com. Visit this group at http://groups.google.com/group/google-appengine?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
