imho: I'd be cautious about AppScale-- you may inherit all the worst
problems
   of AWS and GAE, with few benefits.  One win however-- you'll have
one
   [low level] system to keep running, diagnose, etc.

also imho: I wouldn't wait for GAE/Business or any pre-announced
offering
   from any company: pre-announcements have a tendency to slip.

hope this helps,
adam


On Dec 2, 8:10 pm, Shane <[email protected]> wrote:
> Hi Adam,
>
> Thanks for the response.  One way I was considering approaching the
> issue of keeping the doors open to migrate back to GAE at a later
> date, is to use something like AppScale on AWS.  This way I can be
> more or less on the same path as GAE.
>
> I am really keen to see the GAE for Business plans for non-intranet
> solutions.  This could be an awesome opportunity, however until this
> happens, I just can't rely on a beta product.  It's a real shame
> because it's a lovely system.
>
> Shane
>
> On Dec 3, 8:09 am, Adam Sah <[email protected]> wrote:> I'm a *very* 
> longtime GAE user and former googler and I love it... but
> > for now,
> >    I usually recommend it as an *ingredient* and not as the backbone
> > of the
> >    whole system.
>
> > Specifically, GAE can't support certain libraries (e.g. anything with
> > C/C++),
> >    so you inevitably need another service like AWS, linode, etc.  I
> > use Linode
> >    to run SOLR for us and another PC instance for our offline pipeline
> > (maybe
> >    move to Tasks???)
>
> > GAE makes a nice frontend-- just be prepared for whining about short
> > outages,
> >    limitations, etc. -- OTOH, I love not carrying a pager, instant
> > scaling, etc.
> >    I once did 600 QPS on appengine, effortlessly.  I've run $MM
> > businesses
> >    on appengine, painlessly.
>
> > One thing: if you start on GAE you can easily migrate away, but if you
> > start
> >    elsewhere it's very hard to migrate to GAE.  This argues for giving
> > it a shot,
> >    then re-evaluating (say) monthly.
>
> > For my latest startup, we're based entirely on GAE and it's love-hate,
> > but we've
> >    stuck with it and the issues haven't been big enough to matter--
> > and v1.4
> >    (any minute now!!!) should solve 50+% of the hassles.
>
> > adam
>
> > On Dec 2, 10:03 am, Grzegorz Machniewski
>
> > <[email protected]> wrote:
> > > > Do you know how Zynga's backend is set up on AWS?  I would imagine it's
>
> > > pretty complex so, you have to have the system architects and 
> > > administrators
> > > to manage that.
>
> > > Zynga is using RightScale  as a platform management software 
> > > (http://www.rightscale.com/)... it automates management and scaling of the
> > > application. It's quite impressive solution,
>
> > > Regards
> > > G.
>
> > > On Thu, Dec 2, 2010 at 6:34 PM, Eli Jones <[email protected]> wrote:
> > > > Maybe one of the Google developers will see this and can provide you 
> > > > with
> > > > an informed answer.. but, if it slips through the cracks, you should
> > > > definitely ask during the IRC office hours (or just check #appengine on
> > > > irc.freenode.net, since someone from Google may be there at any time.)
>
> > > > If your app becomes popular on a Farmville scale, you'd have to be 
> > > > prepared
> > > > to automatically start-up new instances and balance load across them.  
> > > > Which
> > > > means you'd need to spend time and energy designing this sort of system 
> > > > up
> > > > front.
>
> > > > Though, if you just plan on cranking up an Extra Large EC2 Instance and
> > > > running off of that (and maybe using SimpleDB on the back end).. and 
> > > > dealing
> > > > with front end scaling issues later.  You would probably be fine.. most 
> > > > apps
> > > > do not become popular on the Farmville scale.  SimpleDB will enforce a
> > > > little database discipline, and, if you do "blow up", you could probably
> > > > figure out how to load balance over multiple static instances until you
> > > > cooked up some sort of automated scaling method.  (Again, this depends 
> > > > on
> > > > you using SimpleDB on the back-end.)
>
> > > > With Appengine, you avoid any sort of systems architecting (just code as
> > > > fast as possible).. but you have to spend time coding ways to deal with 
> > > > the
> > > > 99th percentile outlier Datastore issues.
>
> > > > Maybe there are some people here who have user facing apps who can speak
> > > > about this from experience (I happily do backend processing where 
> > > > retrying a
> > > > failed datastore operation or a failed cold start is no problem.. and 
> > > > there
> > > > is no user on the other side to become frustrated.)
>
> > > > On Thu, Dec 2, 2010 at 1:10 AM, Shane <[email protected]> wrote:
>
> > > >> I know this has been asked one way or another before, but most of the
> > > >> main issues to do with GAE stability seem to have been asked around
> > > >> the end of 2008, early 2009.
>
> > > >> Basically, I have been arguing back and forth with my business partner
> > > >> about whether to use GAE or AWS for the back-end of our social game
> > > >> engine.  I love GAE (Java) for so many reasons, and although it used
> > > >> to be unstable, it's pretty good now.  The main argument in favour of
> > > >> AWS is the fact that AWS has proven itself with multiple games running
> > > >> tens of millions of active users per day.  The obvious pin-up child
> > > >> for AWS is Zynga, with its Farmville peaking at 80+million DAU.  And
> > > >> that's just one of the hugely successful games running on the AWS
> > > >> infrastructure.  Remarkable achievement.
>
> > > >> So, one way or another it's KNOWN to work.  GAE on the other hand
> > > >> doesn't have any examples that I could find doing these sorts of
> > > >> numbers.  Not even close.  So can I trust it?  Is there a single
> > > >> example of a large social game with millions of Daily Active Users,
> > > >> using GAE?
>
> > > >> I look forward to your thoughts, but please also note, this is not
> > > >> intended to start any sort of flame war.  I love both systems, but
> > > >> both have their positives and negatives, but I'm about to make an
> > > >> architectural decision that likely won't be undone moving forward.
>
> > > >> Regards,
> > > >> Shane
>
> > > >> --
> > > >> 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]<google-appengine%2Bunsubscrib
> > > >>  [email protected]>
> > > >> .
> > > >> For more options, visit this group at
> > > >>http://groups.google.com/group/google-appengine?hl=en.
>
> > > >  --
> > > > 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]<google-appengine%2Bunsubscrib
> > > >  [email protected]>
> > > > .
> > > > For more options, visit this group at
> > > >http://groups.google.com/group/google-appengine?hl=en.
>
> > > --
> > > Grzegorz Machniewski

-- 
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