Public Cloud is Awesome, and We Found A Way To Make It Cost Less!

By Mike Westerfield, Enterprise Architect

Spot changes the economics of the Public Cloud

When I was first introduced to Spot (NetApp acquired Spot in mid-2020), I have to admit I was a very skeptical.  The offer of saving from 70% to 90% on cloud compute spend sounded too good to be true and offered a significant departure from other cloud cost analytics and optimization solutions.  The Spot offerings are SaaS-based, and the way they charge for their offerings is a percentage of the savings they deliver to you, so it is a no-risk investment for our clients.  Spot makes money when you save money on your public cloud bill.  They call their offerings “budget freeing” solutions.

Spot – It sounds too good to be true…

You might be asking yourself if you are giving up performance or accepting a lower SLA by utilizing the Spot service, the answer is no.  When using the Spot automated cost optimization offering, Spot guarantees the same SLA (99.99% uptime for EC2 on demand instances).  You might also be asking yourself if you will gain any savings if you use Reserved Instances and Spot instances from your public cloud provider.  In this case the answer is also a surprising yes.  Spot has a great chance of still reducing your monthly public cloud bill!  I bet one more thought that is going through your mind is, “Spot is going to make me move off the public cloud to some private cloud to get these benefits.”  That is not the case, Spot’s sole focus is to cost-optimize your use of your current Public Cloud provider!

Spot – What exactly are the offerings?

Spot offers 3 ways to save your company money:

  1. Elastigroup leverages your auto-scaling workloads and transitions them from using expensive on-demand instances to less expensive spot instances. Elastigroup simplifies infrastructure and reduces costs for web and other scaling workloads using analytics and automation.  Elastigroup features:
      • Dramatic Cost Savings
      • Ensured availability for production workloads (99.99% just like On Demand!)
      • Intelligent scaling
      • The ability to optimize without having to change anything
  1. Ocean leverages your containerized workloads in a similar manner and moves them from on-demand to the cost-optimized spot instances. Ocean frees operations teams from the overhead and complexity of managing container infrastructure.  Ocean provides:
      • Automated infrastructure management
      • Container level visibility
      • Cost control and optimization
      • Provides container driven, predictive autoscaling to ensure pods and tasks always run on the right blend of compute infrastructure
  1. Eco optimizes your use of AWS reserved instances (RIs) and Savings Plans. Eco analyzes your current and projected compute consumption, creating and executing an optimal purchasing strategy of lower-cost, 3rd party RIs (sourced in the AWS RI Marketplace) along with the right blend of standard RIs, convertible RIs as well as Savings Plans when appropriate. Eco will offload any unused RIs in the AWS RI Marketplace ensuring that you always enjoy long-term pricing with minimal financial lock-in. As of this blog, Eco is not yet available for MS Azure or GCP, so contact Dasher as it will be coming soon!

You might be questioning the use of spot instances for production workloads, but this is no issue for Spot.  By utilizing Spot’s data analytics and automation, Spot can guarantee workload uptimes using available spot instances.  What happens if there aren’t any spot instances available that meet the workloads requirements?  Spot will fall back to on-demand instances until spot instances become available again.

Not sure if you have enough instances to take advantage of this type of savings? Spot has you covered there too.  The final offering from Spot is called Cloud Analyzer and is free to use today!

Cloud Analyzer will look at your environment and determine how much the solution can save you without having to on-board your workloads to the various automation solutions already mentioned (Elastigroup, Ocean, and Eco).  Cloud Analyzer can be connected to your Public Cloud in environment in just minutes.

There’s a number of case studies showing the effectiveness of this solution from companies such as Ticketmaster, Duolingo, Netapp, Samsung and others.  Take a look at them here:  https://spot.io/case-studies/

As of this writing (February 2021) Spot offers its full suite of products to AWS users.  Elastigroup and Ocean are supported for MS Azure and GCP today. Spot is working hard to complete the full suite of solutions for MS Azure and GCP, so if you need these please reach out to Dasher to discuss the roadmap.

Summary and Conclusion

If you’re currently spending more than $70K-$80K a month in cloud spend and have workloads that leverage auto-scaling, containers, or reserved instances let us know and we can provide a deeper dive on the Elastigroup, Ocean and Eco solutions.

If you have further questions about Public Cloud environments and using Spot to optimize your costs, or would just like to talk about Public Cloud with our engineering team, feel free to reach out via our Contact Us page.

 

 

 

 

 

This post is powered by Mix Digital Marketing