On Thu, Feb 17, 2011 at 7:15 AM, Michael Abner <mike.ab...@gmail.com> wrote: > In this situation I would think you'd want to upload to s3 directly and the > fire off a background worker to go and do the resizing for you. That would > keep your web dynos available to serve your requests.
Exactly right. Uploads straight to S3 from the browser are fantastic, you should use them if you're doing any kind of substantial upload processing. Moral of the story: never, ever block the web process for long (> 1 second) periods of time - there's always a better way. Follow this practice and you and your users will always be happy campers. I certainly extend my apologies to you, Christos, and anyone else inconvenienced by the incorrect docs. I've tweaked the performance concepts doc to remove some of the more glaring inaccuracies, which will hopefully tide us over until we can rewrite the whole thing to cover the broader array of concurrency models we can handle. Adam -- You received this message because you are subscribed to the Google Groups "Heroku" group. To post to this group, send email to heroku@googlegroups.com. To unsubscribe from this group, send email to heroku+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/heroku?hl=en.
