As usual, the new services are available on a pay as you go basis.
The Auto Scaling feature is being bundled with the Monitoring service (CloudWatch) - both in terms of features and prices. Pricing for these services is based on the number of instances to monitor, with automatically provisioned instances being charged at the standard rate. Detailed information can be found on the CloudWatch product page. As an example, monitoring 10 instances would cost about $108 (USD) per month.
Load Balancing is available across availability zones - you can create a new Load Balancer via the updated API and then map the balancer to your domain via DNS. As usual, the Elastic Load Balancing product page will give you all the details you need. There is also a pricing example:
As an example, a medium-sized website running on 10 Amazon EC2 instances could use one Elastic Load Balancer to balance incoming traffic. If the Elastic Load Balancer ended up transferring 100 GB of data over a 30 day period, the monthly charge would amount to $18 (or $0.025 per hour x 24 hours per day x 30 days x 1 Elastic Load Balancer) for the Elastic Load Balancer hours and $0.80 (or $0.008 per GB x 100 GB) for the data transferred through the Elastic Load Balancer, for a total monthly charge of $18.80. Partial hours are billed as full hours.
For us, these latest updates mean that we'll give load balancing a try with the hope of replacing our current solution. For now, at least, we'll leave our custom monitoring and scaling in place.


1 comment
Comment on Amazon CloudWatch, Auto Scaling and Elastic Load Balancing by AndrewBoldman
June 4th, 2009 @ 22:43 – permalink
Leave a reply
You can use Markdown in your comment as well as plain HTML. You can use
<filter:jscode lang="ruby">and</filter:jscode>tags to surround code blocks (supported languages are css, html, javascript and ruby). Your email address will not be published.If your comment doesn’t appear immediately after posting it could have been marked as spam. Don’t worry: we regularly check for and approve incorrectly filtered comments so you shouldn’t have to wait too long for it to be shown.