Back to Blog

Jacob Ben-David

Multicloud Reserved Instance Management Made Easy with Turbonomic v6.4

Turbonomic 6.4 includes numerous enhancements and new capabilities for our public cloud customers. Our platform is designed to assure application performance with Application Resource Management (ARM) at any stage of their cloud journey. From cloud migration planning and modeling to IaaS or PaaS services to AI-powered workload optimization of public cloud workloads on Azure and AWS. At any stage, Turbonomic is designed to unlock the true potential of the cloud by carefully balancing application performance with optimal efficiency by leveraging every discount mechanism offered by the cloud provider.

The Reserved Instance Savings…and Pains

One of the most cost-effective discount mechanisms on the cloud is Reserved Instances (RIs). Offered by both Azure and AWS, RIs discounts are based on 1-yr or 3-yr commitment to consuming resources at a significantly lower price when compared to on-demand, up to 75% in some cases.

Customers who leverage RIs, face two main crucial and complicated decisions related to RIs. These decisions are not a one-time effort and must be done continuously as the cloud footprint and workloads' demand change.

The first fundamental decision is how much reserved capacity to purchase. This decision includes multiple data points that are influenced by the workload types, their uptime, and their demand patterns.

The second critical challenge is how to manage and ensure that the actual utilization of purchased RIs will be high, closer to 100% as possible. The problem with achieving this is rooted in the long-term commitment of RIs. As workload demand fluctuates, they need to be resized, and by doing so, some RIs will be left unused.

Our customers told us that before Turbonomic, they spent countless staff-hours, and despite their efforts, they could not attain satisfactory utilization level of their RIs.

AI-Powered Multicloud Reserved Instance Management

We are excited to share that in Turbonomic 6.4, we have expanded our Reserved Instance management capabilities to Microsoft Azure. Turbonomic will now generate AI-Powered Reserved Instance purchasing and RI Inventory-aware VM scaling, to maximize both RI utilization and RI coverage for Azure and AWS.

Turbonomic will continuously assess the utilization of all workloads on Azure (per region) and will generate accurate RI Purchase recommendations that will help increase RI Coverage.

As part of our UI enhancements, we added visualization to our buy action details to help users understand the reason for the purchase action. The visualization consisted of a graph showing the consumed resources compared to the suggested capacity to buy for the VM group. The window will also provide details on the expected costs, savings, and projected utilization of the new RIs.

Turbonomic also provides a robust simulation and modeling engine to help customers determine the best RIs to buy for workloads running on the cloud. A typical example is comparing the required investment and potential savings of different RI purchasing terms or RI types.

It is also important to note that Turbonomic will provide RI purchasing recommendations for VMs migrated to the cloud with Turbonomic's Cloud Migration planning.

In Turbonomic 6.4 we also added support for Azure SQL RIs recommendations for Microsoft SQL Servers migrated to Azure SQL DB Managed Instance Service. The cloud migration plans to Azure also support Azure Hybrid Benefits for SQL and Windows workloads, and when combined with RIs can offer up to 80% discount.

 

In parallel, Turbonomic Application Resource Management will continuously generate VM scale actions that are fully aware of the existing RI inventory. RI-Aware scaling means that when workloads require resize, Turbonomic will first assess the current RI inventory and will size to an available RI, resulting in optimal performance at the lowest cost. The same applies to idle workloads, even if already covered by RI, Turbonomic will precisely size them to a best compute offering that will provide better savings and performance. The new size could be a different RI, and the current RI might be used by a different VM that could benefit from it.

Most cost optimization solutions use different analysis engines for RI purchases and workload rightsizing; most will rightsize without considering existing RI inventory. Turbonomic is the only AI-powered platform that uses the same AI engine to generate RI purchasing recommendations, and automatable RI-aware rightsizing actions.

 

Conclusion

Reserved Instances are one of the best ways to control and reduce the spend on Azure and AWS, but with significant savings come significant overhead. Effectively managing RIs is a hugely challenging endeavor with a single cloud, and it is beyond human scale when adopting a multicloud strategy.  Using Turbonomic to manage your public clouds will guarantee application performance at the lowest cost possible while being business compliant.

Application Resource Management offers continuous sizing of workloads based on application demand, coupled with accurate and trustworthy RI purchase recommendations that help organizations stay within budget while unlocking the elasticity and operational agility of the public cloud. Give our platform a chance and see for yourself.