AWS

Archana Kesavan of ThousandEyes speaking at NANOG75 reports that network traffic between AZs within a single region is generally “reliable and consistent,” and that tested cloud providers offer a “robust regional backbone for [suitable for] redundant, multi-AZ architectures.”

ThousandEyes ran tests at ten minute intervals over 30 days, testing bidirectional loss, latency, and jitter. Kesavan reported the average inter-AZ latency for each tested cloud:

Within the four tested regions in AWS, they found:

Kesavan’s slides and video are online.

If you’re running infrastructure and applications on AWS then you will encounter all of these things. They’re not the only parts of a network setup but they are, in my experience, the most important ones.

The start of Graham Lyons’ introduction to networking on AWS, which (though the terms may change) is a pretty good primer for networking in any cloud environment. Though cloud infrastructure providers have to deal with things at a different later, Graham’s post covers the basics—VPCs, subnets, availability zones, routing tables, gateways, and security groups—that customers need to manage when assembling their applications.

What we do first is we build very simple foundational building block services … we will build the simplest possible service that you could think of.

The next thing we do is we encourage an open marketplace within Amazon so individual teams can use, optimize, and extend these basic services.

We use our individual [teams] as a test lab to experiment on better ways to do things, and when we find something that seems to be working, we look for ways to [grow it and use it more] broadly.

Then, on the value of open repos:

We want this ecosystem of learning from each other because we are all leveraging each other’s web services. We have these hardened contracts, it’s incredibly high leverage to be able to go and see how someone else used a web service quickly, and rip a piece of their code—steal it—and make use of it for your own purposes.

Andy’s slides are at Slideshare.

Amazon calls it CloudFront, and it costs $0.17 – $0.22 per GB at the lowest usage tiers. It seems that you simply put your files in an S3 container, make an API call to share them, then let your users enjoy the lower-latency, higher performance service.

Their domestic locations include sites in Virginia, Texas, California, Florida, New Jersey, Washington, and Missouri. Internationally, they’ve got Amsterdam, Dublin, Frankfurt, London, Hong Kong, and Tokyo covered.

ProgrammableWeb pointed out an InformationWeek story that claimed 28% of Amazon’s sales in early 2005 were attributable to Amazon affiliates. And C|net claims Amazon now has 180,000 AWS developers (up from the 140,000 Amazon was claiming about a year ago).

(Note: not every Amazon affiliate/associate is an Amazon Web Services (AWS) developer, but Amazon hasn’t shared more specific numbers.)

These slides, from Amazon’s AWS developer relations team explain a lot about what AWS is.