I think this should be great for my new "high confidence" put
architecture changes (original thread below).

If I'm reading this right, I can set up a separate High Reliability
application that will handle the put().

These puts are much lower volume than other client activities, but I
really need high reliability for them.

So architecture at a summary level is:

1) Client initiates the put to MS application.
2) MS application launches a task queue task to HR application.
3) MS replies back to client that HR task is launched.
4) Client pings HR application to confirm write (showing appropriate
dialog).
5) HR task queue puts record.
6) Client now gets HR confirmation to ping and shows confirmation
dialog.

If I've missed something please advise. The higher cost and higher cpu
cycles for HR (link below) are not issues as only a subset of
critical, lower volume puts get HR costing.

If I've not missed anything, then pleased to say in advanced (before
refactoring code) "Thanks Google".

Links:
Previous write architecture thread:
http://groups.google.com/group/google-appengine/browse_thread/thread/c0e44d36d54384a3

MS vs. HR performance testing (thanks kaekon):
http://groups.google.com/group/google-appengine/browse_thread/thread/5fc3b6a4366de62f

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine?hl=en.

Reply via email to