Amazon Costs: Kubernetes to the Rescue

February 27, 2020

How much does an AWS customer pay for a specific service? DarkCyber knows that the taxi meter approach makes sense for the company that owns the medallions, the taxis, and possibly a few important people.

The taxi meter method in the cloud is a bit of a mystery — until the invoice arrives. “The Story Behind My Talk” explains a process which, according to Tuananh, can “reduce EC2 billing up to 80%.”

How does this money slip from the massive tractors of the Bezos bulldozer?

The article explains:

I started with a managed GKE at first using their free $300 credit, to learn the basic of Kubernetes; but then later on use kops to setup a production cluster with AWS.

Some of the changes I did for the production cluster is:

  • Setup instance termination daemon to notify all the containers + graceful shutdown for all the apps.
  • Setup multiple instance groups of various size and availability zone, mixing spot instances with reserved instances. This is to prevent price spike of certain spot instance group; and minimize the chances of all spot instances going down at the same time.
  • Calculate and provision a slightly bigger fleet then what we actually need so that when there were instances shut off, there won’t be service downgrading. Because spot instances are so cheap, we can do this without worry much about the cost.
  • Watch to see if there were scheduling failure to scale the reserved groups.

The payoff according to the article:

“The overall cost saving for EC2 was around 60-70% because we need to mix reserved instances in and provision a little higher than what we actually need. We were very happy with the result.”

Will AWS reconsider how it deals with EC2 billing? Does a bulldozer need diesel fuel?

Stephen E Arnold, February 27, 2020

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta