Access and Feeds

Serverless Computing: Re-engineering the Cloud to be ‘Serverless’

By Dick Weisinger

Serverless computing doesn’t mean that there aren’t servers.  It means that the infrastructure needed to be provisioned in order to run computing jobs is automatic.  The focus is on the application or service and not on the server hardware or operating system.  Serverless systems try to correctly match the RAM, CPU and bandwidth requirements for the computing task and to scale as needed to match the requirements.  All the scaling, provisioning, and patching is handled for you.

Amazon was the leader with the serverless concept, but now all the large cloud vendors offer it in some form.  There is Amazon AWS Lambda, Microsoft Azure Cloud Functions, Google Cloud Functions, and IBM OpenWhisk on Bluemix.  Other smaller vendors like Terraform, Fission and Fn also have offerings.

It’s not just services.  Amazon has, for example, a serverless version of its cloud database Aurora.  The database can scale to 64TB of storage based on the requirements of the application.

Increasingly we can expect a variety of other cloud services to be reengineered to use a serverless approach. The serverless trend should lead towards greater utilization and efficiencies.

 

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Leave a Reply

Your email address will not be published.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

*

thirteen + twenty =