Hi,

Can the Google team please shed some light on the issue of multiple
datacenters in various geographical locations pertaining GAE? I would
like to understand what characteristics and what level of guarantees
to expect from a GAE app, and didn't find any docs or posts on this...

1. Is static content delivered by a geographically distributed CDN?
Would make sense for latency and international bandwidth costs.

2. Does GAE launch VMs of a single app in multiple datacenters? Same
as above - latency would be quite bad for some geographical locations
without this.

3. If VMs are distributed, what's the deal with memcache? If it's
distributed, does/can it split into distinct partitions? How does that
affect atomic operations, such as increment, or things like grabTail?

4. If there's geographical distribution, how does it affect Datastore
performance and latencies?

My current (disappointing) impression based on latencies is that GAE
VMs are not geographically distributed, based on the latencies. But
would be good to get some clear information on this.

In general, it would be quite useful to be able to monitor VMs better,
eg. how many VM launches took place, where. This would provide more
insight into the operation and performance of the app... (This may
also be a stepping stone toward the "pay to keep some VMs warm" topic,
an often requested feature - at least we could see )

Thanks,
  Viktor

-- 
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 [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-java?hl=en.

Reply via email to