Scaling
When scaling becomes available how will added instances be spread across zones in Amazon's IaaS? None of the main IaaS vendors in the market today are without their downtime stories, but how much fault tolerance can we expect going forward?
I am not an expert on Amazon's IaaS. In fact, most of us using your service probably aren't which is why we're here.. and why I am asking this question here.
Discussions are closed to public comments.
If you need help with AppHarbor please
start a new discussion.
Keyboard shortcuts
Generic
| ? | Show this help |
|---|---|
| ESC | Blurs the current field |
Comment Form
| r | Focus the comment reply box |
|---|---|
| ^ + ↩ | Submit the comment |
You can use Command ⌘ instead of Control ^ on Mac
Support Staff 1 Posted by friism on Aug 20, 2011 @ 06:23 PM
Currently AppHarbor application servers are located in a single availability zone. This makes an application with multiple instances durable in case of an individual server failure, but not in case an entire availability zone becomes unavailable.
Deploying applications on different servers hosted in different availability zones is an option we are actively considering. To get the most of this, the scenario involves running loadbalancers on hot standby on multiple availability zones.
Ensuring maximum uptime and reliability for you application is at the top of our minds, and we're constantly working to improve the durability of the AppHarbor platform.
Let us know if you have other questions or if you want to trial having your application deployed to multiple instances.
Michael
friism closed this discussion on Aug 20, 2011 @ 06:23 PM.