What are Google's long-term plans for Google App Engine? | Hacker News Reader

What are Google's long-term plans for Google App Engine?

47 pointsrnc00015 comments

GAE as a platform is really great (good scalability, no maintenance overhead, low latency in most cases) but the extremely slow pace (compared to Amazon Web Services) for introducing new APIs and maturing the existing (ex.: increasing quotas and charging for the Search API, Prospective Search) makes me wonder: is Google really investing enough engineers to mature its platform? Will it pull the plug and stop offering it? Is the Premier support option really helpful? What are it's long-term plans? Where's the roadmap? These are very important questions considering it's PaaS and you'll be deeply committed to it "by design" once you Go Google.

View on HN

I think no discussion of GAE's future should ever miss these very important options:

http://appscale.cs.ucsb.edu/index.html

http://www.jboss.org/capedwarf

So even if Google shuts down GAE, you could transition without too much trouble.

Regarding the question itself: GAE is just a friendly face on top of Google's internal systems, and Google is deploying all of their new stuff onto it directly for ease of maintenance. GAE is also well charged so that Google is making a profit off it, so they have no reason to shut it down.

The big issue is if you are a small developer who cannot shell out huge amounts for usage charges. Google is targeting GAE more towards Corporate and the kind of budget a large business has, rather than towards a guy in his basement doing hobby work. The charges are still pretty low though, but it's not unlikely that they may raise in future.

Conclusion:

Deep pockets? Go for it, you can't really find a better scaling option for the cost.

Want free hosting? Not a good choice, look elsewhere.

Indy dev who wants to pay very little and is willing to optimize caching to lower costs? Solid choice, you can get the charges very low for even high demand sites. Keep a sharp watch on billing trends, and test out your site on AppScale / CapeDwarf to make sure you can transition off quickly if billing gets out of hand.

"Want free hosting? Not a good choice, look elsewhere."

What are some better alternatives? And why are they better?

I've used GAE for some toy apps and a few websites with low traffic. It works great.

I'm not sure which criteria he's using for "better", but these guys give you a lot, plus flexibility for nothing:

http://www.appfog.com/

https://openshift.redhat.com/app/

I don't understand replicating the restrictions, challenges, and proprietary feel AppEngine makes towards normal application style (compared to something like Heroku).

Lots of internal apps at google are written on top of appengine. For example the android developers portal is written on top of it. So i woudnt worry too much.

Good point. But still, it's hard to believe that it's taking years for Google to add to GAE a mature Search API and support for multiple datacenter locations, which are things Google does better than anyone.

Developers outside of Google are not really ready to cope with the daily reality of being inside Google. Things that Google does "easily" like search and multihoming with replication and failover and so forth are actually tremendously difficult. App Engine customers are not willing to embrace the limitations that Google's architecture dictates (indeed, App Engine customers are constantly complaining about the few such very minor limitations to which they are already exposed.)

I use GAE on a few b2b, process oriented webapps of my own and also do GAE consulting for large customers.

I've met a few times with their evangelists/sales/support people in my country. They have always been very enthusiastic and are pushing very hard for adoption by large corps and universities. I don't have the impression they will pull the plug but... you never know.

As for the Premier support option, the few times I needed it they were very responsive and helpful; I do recommend it if possible.

To a solo entrepreneur like me, GAE is a great option and I can't imagine myself managing servers and a full stack, even if it is just building/deploying an AMI. GAE is just too easy.

This question applies to all proprietary platforms. By tightly integrating your application or service, you're pretty much at the mercy of the PAAS provider.

To avoid vendor lock-in, check out the OpenShift platform [1] from Red Hat. It's open source [2] so you can set it up on your own servers if you felt Red Hat weren't providing the service you required.

[1] https://openshift.redhat.com/community/paas

[2] https://openshift.redhat.com/community/open-source/download-...

Lack of info from Google is a big problem for me. Who knows what the memcache quota is, for example? [1] They drastically raised prices with little warning. GAE doesn't seem professional enough to depend one's business on it, unless it's a relatively tiny app.

[1] http://stackoverflow.com/questions/8276292/google-app-engine...

Understandably for a PAAS they tell you to treat memcache as if it wasn't there and make no promises as to the quota.

The reason is that they can twiddle things around behind the scenes without anybody architecting their app around a particular memcache performance fingerprint. If you think about it, that's how you are supposed to treat something ephemeral as memcache anyway, best effort availability, not to be relied on.

I wouldn't worry about this too much. Their datastore being a sort of blackbox is more of a problem. Check out the Khan academy dev blog, this is not a bottleneck.

That's not a very useful service if the memcache can disappear. It's difficult to properly provision a service when the average case costs X and the worst case costs 1000X because the cache layer is "ephemeral". When the cache is working your cost utilization will look like shit and your financial management is going to ask you to cut costs. Then you'll do that and when the cache suddenly disappears, you'll suffer an outage because you are out of capacity.

So I think it's not very reasonable to offer no-SLA components of a hosted platform. It's either there with some assurances, or it might as well not exist at all.

(Khan Academy dev here.) The datastore being a black box is a slight concern to me personally, but the really great tradeoff is letting Google handle the responsibility of keeping our persistence layer, among many other services, up and running. They're quite good at doing that.

Great conversation. Our best info is that Google is committed to GAE and so are we. AppScale can be the "on-ramp" to GAE and also your insurance policy if "lock-in" is a concern. Run AppScale on top of EC2 or Eucalyptus (or others) for portability across clouds.

If I sign up for the Premier support, would they answer these questions?

Reply on news.ycombinator.com