Hi Robin, This is a great idea, in fact we already replicate data into more than one datacenter. At the moment, one datacenter is primary for a given app and data is replicated in the background to other datacenters. There is a small delay in replication, which we are in the process of shortening. When we do fail over, we put the primary datacenter into read only mode while replication completes, then we select the new primary datacenter. In the future we may no longer need to have just one datacenter be the primary.
This is covered in more detail in a recent blog post: http://googleappengine.blogspot.com/2009/09/migration-to-better-datastore.html Thank you, Jeff On Mon, Sep 28, 2009 at 1:03 PM, Robin B <[email protected]> wrote: > > Here is an idea for better datastore fault tolerance: > > Why not have >1 datastore available to each application. When Google > needs to do maintenance, they put at most 1 datastore in read-only > mode, and applications can revert to 1 or more alternate datastores > until the maintenance is completed, then apps can use their task queue > to resync their own data. > > A person can accomplish this right now using AWS SDB when BigTable > becomes unavailable, but sending that data over the internet is not > ideal. > > One of the big selling points of GAE is scalability, fault tolerance, > and uptime, but having no persistent write access for any period of > time prevents people from achieving those ideals. > > Are temp/failover datastores a plausible idea? > > Thanks, > > Robin > > > --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
