My apologies if this is an FAQ I just haven't noticed, or something
that's mentioned right at the beginning of one of the [many]
references I haven't managed to dig through yet (especially if it's in
Bruce Slatkin's Google I/O session that's so basic--I've had that page
opened in a tab for close to 6 months now, and I never seem to have an
hour I can justify taking to watch).

Anyway, the most important basic point of all the <s>limitations</s>
opportunities we embrace is for scalability, right?

I just recommended to someone that he upload a few thousand records to
google's datastore and then run the queries against them that he
expects to be the most common/slowest.  I think the size of the
resultset might affect his query performance, but not the number of
records in the data store. I assumed that, if a few thousand records
work okay, then so would a few million.

In the middle of writing that, I realized that my assumption was
pretty much completely unjustified.  That's the impression I've
gathered from everything I've read, but I don't think I've seen any
sort of official statement about that.

I don't expect any sort of iron-clad "Your query will run in O(1)
time, ignoring the size of the datastore" guarantee, or anything like
that.  But isn't that more or less the goal?

Or should I also start thinking about ways to archive/data warehouse/
etc, the way if I were running an RDBMS?

Thanks,
James

--~--~---------~--~----~------------~-------~--~----~
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