I have to concur with Corby on this ... I was initially attracted to the end-to-end solution that GAE offered integrated as it is with Eclipse.
In practice I have found it very frustrating to develop on GAE because of: - the class white list and 3rd party framework usage (i.e Spring) - the truly frustrating metering exceptions that the Datastore and Servlet container throw, coding around those is painful... you have to invent your own "transaction" compensation... - the lack of any Java based bullk load facility for DS (I had to write my own) (which is when you trip up over the metering exceptions...) - the limitations of mapping BigTable under JPA/JDO - the paucity of debugging on the production server (the log WUI is woeful for this) We have reached the stage with our app where we will probably move it off GAE and onto EC2 where despite the additional work involved in dealing with AMI lifecycle vs .war we are more in control of the underlying environment (although I am sure there are a similar, but different, set of hiccups and hurdles awaiting us ...) - Larry On Feb 24, 10:17 am, Corby <cep...@gmail.com> wrote: > On Feb 24, 10:39 am, A1programmer <derrick.simp...@gmail.com> wrote: > > > Are there any good alternatives to Google App Engine out there? > > > I am finding that I cannot rely on GAE for a production quality > > environment. > > I am very excited about the future of Google App Engine, but I am in > the same boat, Derrick. > > I am working with a startup that tried to get up and running on Google > App Engine, but we have since migrated to Spring Cloud Foundry, which > runs on Amazon's EC2 architecture. It is a heavier-weight environment, > and the monthly costs are higher, but the development costs are much, > much cheaper, and we have far fewer unexpected Production problems. > > > There's a huge attraction to the tiny resource requirement to maintain a WAR > > file deployment versus the large requirements of a EC2 style VM. > > I agree, but for me there is an even larger attraction to having > predictable behavior in our Production environment, and we can't > currently get that from GAE. Some people are happy with BigTable as a > general persistence solution, but we are not. The whitelisted > classloader makes every attempt to integrate with third-party > libraries an adventure. And the nature of GAE's on-demand provisioning > makes it difficult to guarantee the response times we want for our > customers. > > The crushing blow to our morale is code that runs perfectly in our > development environment, but breaks when deployed to GAE. > > I have a lot of faith in the Google team, and I can't wait to see what > GAE looks like a year from now. But right now, I can only recommend it > for the simplest of internal apps, nothing client-facing. -- You received this message because you are subscribed to the Google Groups "Google App Engine for Java" group. To post to this group, send email to google-appengine-j...@googlegroups.com. To unsubscribe from this group, send email to google-appengine-java+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/google-appengine-java?hl=en.
