Back to Blog

Eric Wright

Intelligent Autoscaling for Public Cloud with Turbonomic 6.3

Autoscaling applications have become a favorite feature in public cloud environments.The challenge is that they are built on the premise of scaling up rapidly at an arbitrarily specified threshold, and scaling down slowly to maintain availability. While this does achieve scaling of applications, there is a better way! On that note, we're incredibly excited to deliver continuous intelligent cloud autoscaling with availability-set awareness. This enhanced capability delivers consistent scaling while simultaneously assuring application performance and resilience in AWS and Azure environments. 

Intelligent Cloud Autoscaling

Turbonomic 6.3 provides the most comprehensive approach to continuously and intelligently sizing your scale-out IaaS applications on the public cloud. This is a result of enhancing our existing scaling functionality to include consistent scaling of instances and SKUs within scale-out configurations to optimize performance and cloud spend while ensuring your application availability.

custom-cloud-costs

The questions that today's application architects and cloud operations teams are facing include: 

  • When do you scale the application out?
  • How do you choose the initial configuration and sizing?
  • When do you scale up versus out?
  • How do you achieve performance without runaway cloud costs?

By using Turbonomic intelligent autoscaling, the application demand is met dynamically and continuously by knowing when to scale up and when to scale out to get the optimal use for both performance and cost. As demand fluctuates, the scaling configuration is updated automatically by Turbonomic to meet that demand and can be automated nearly every step of the way. This can also hand off to your existing configuration management platform if desired.  

On Azure, changes made to running instances that require restarts are fully automated and completed using a rolling restart across the availability set with confirmation of each change, to ensure availability during any configuration and optimization processes. 

consistent-resize-azure

AWS configuration changes are done at the launch configuration to maintain consistency with existing scaling processes and restarts can be triggered separately to initiate the updates. To learn more about Azure and AWS IaaS Solutions, check out these eBooks: AWS eBook, Azure eBook

consistent-resize-aws

Top that off with the fact that these dynamic scaling decisions are also fully aware of reserved capacity on public clouds so that you get automated matching of running workloads to available reserved instances and also planning features to help you buy reserved capacity intelligently to match real application demand based on both real-time and historical utilization.  

Bring Consistent Scaling On-Premises

Not only our public cloud customers are gaining the benefit from intelligent autoscaling. The same features can also be brought on-premises for use-cases such as production and disaster recovery application groups to ensure that changes to scaling in production are reflected automatically in your DR/BCP environment.   

Aligning test and development, production and QA, and of course all being done with an automation-first approach that Turbonomic has built its foundation on since the dawn of the company. This is just the beginning of what you can unlock thanks to the consistent scaling features that have been added to the policy framework in the Turbonomic 6.3 release. 

We’ll be digging into these capabilities and more during the Accelerate and Optimize Your Hybrid and Public Cloud Adoption Strategy webinar, on March 13, 2019. Join us to see these new capabilities in action with both live demos and an interactive forum with Turbonomic experts.