After initial testing, I already love transactions, started generating some gotchas of my own, going to share them when I finish testing/modifications (for newbies to transactions like me)
Before going deeper into transactions, I want to ask some questions: - Is there a downside of using transactions cost/performance wise? ( retries=10 should potentially 10x increase the cost/performance, but is there anything other than that, something like "transactional costs are reflected as 3x to appengine costs"(made up) etc. -- or any overhead other than retries ) - Is there any downside of using xg=True ? Also something like a "dominant transaction" would have been great For example a pipeline routine could be computing 500 operations at once, but a small transaction might take priority and cause the re-computation of 500 tasks, increasing computation costs / slowing down the pipeline etc. -- 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. For more options, visit https://groups.google.com/groups/opt_out.
