0

Lets say I have existing infrastructure for web server(s), application server(s), and database servers.

Now I want to take advantage of some of Amazon's Web Service offerings. Lets take RDS for example.

Am I seeing this correctly: my existing code would connect to and pull data from RDS, across the internet, and then proceed to use the data as it does today? Effectively doubling the HTTP request/response cycle?

In graphical form:

double request overhead

Same question applies to integrating with (say) S3, or DynamoDB.

Corollary question: Presuming the above is undesirable, does not using EC2 (aka Amazon hosted servers) preclude using other AWS offerings?1

1: since EC2 has direct access to most if not all other services, without going through the public internet

Improve this question
3
  • 1
    In most cases, RDS is only a good idea if you're using EC2 for hosting. Latency is a killer for something like an SQL-backed web app - you want to be in the same network. Stuff like S3 can be used from anywhere.
    – ceejayoz
    Feb 28, 2013 at 0:59
  • @ceejayoz: Are there other AWS offerings that have the same "use from anywhere" characteristic as S3?
    – Roatin Marth
    Feb 28, 2013 at 14:17
  • Most of them, including RDS, can be used from anywhere. CloudFront, Simple Email Service, Elastic Transcoder, etc. are commonly used for non-EC2 stuff.
    – ceejayoz
    Feb 28, 2013 at 14:21

1 Answer 1

Reset to default
0

You would use an AWS service because:

  1. The end user accesses AWS directly and not via your hosted server. For example, you could serve dynamic content from your hosted server, but static content from S3 (with the end user accessing S3 directly).
  2. Using AWS buys you something despite the additional transfer cost. For example, it may be prohibitive for you to store a petabyte of data yourself, and so you use S3 instead.
Improve this answer
3

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .