… As A Service – SAAS, PAAS, IAAS

Thanks to nkhumpreys I was introduced to the three “.. as a service” concepts from cloud computing.  Even though I have been using the cloud I had never heard the terms before but they classified the tools I had been using. Here’s my overview.

SASS – Software As A Service

The top of the stack where the user interacts with an application which handles the processing and storage of data. Like a desktop application you access through a web browser.  The user is not usually developing code.

Examples: Google Docs, Dropbox, SalesForce.com

Typical Billing: on a monthly subscription

PAAS – Platform As A Service

A software platform is provided which abstracts away some of the elements of the cloud – normally the scalability elements. This allows a developer to write a piece of code that processes a job (e.g. output a page of a website) and not worry about how many concurrent instances there are. This is great for services that might have unpredictable demand or where there are bursts of activity – such as algorithm that runs for 5 minutes every hour.

Generally the platform makes restrictions on the type of languages of the software and the code must be written in a way so it can execute in parallel – think problems that would work under Map Reduce.

Examples: Google App Engine, PiCloud

Typical Billing: based on resource usage at a granular level – e.g. amount of bandwidth or number of CPU cycles used.

IAAS – Infrastructure As A Service

The renting of a piece of infrastructure for a short period (e.g. a server) in small increments (often hourly).   This reduces the need to keep spare servers to deal with load fluctuation – new servers can be rented on an hourly basis when required.  This provides full access of the instance server including shell access and installed libraries.

Examples: Amazon EC2Rackspace Cloud

Typical Billing: per instance per hour

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.