By Kiersten Putnam | Trace3 Senior Innovation Researcher
Over the past few years, our Innovation team has noticed an uptick in conversations around cloud costs and the best ways to optimize them. We’re noticing many enterprises expanding their FinOps teams, tasking them with tracking all cloud costs and making sure everything is cost-effective. However, as their scope grows, it’s important they have the support required to take advantage of optimization opportunities. This can be tough because while FinOps teams find the opportunities, the cloud engineering teams execute the optimizations. These engineering teams have their own priorities, such as developing fast and creating new IT capabilities, without the demand of formal processes or cost management. So, it's all about finding the right balance between free innovation and resource optimization.
While cultural alignment of these teams is essential for this balance, it is also important to challenge the current processes that have paved the way for cost optimizations. In our research, we have come across emerging solutions that are challenging these processes by making cost optimization a natural part of engineering functions. We believe this has the potential to shift FinOps into Engineering Optimization and thus, we are calling this out as one of our major themes for 2024. If you want to learn more about other themes we are tracking for 2024, continue following our Trace3’s Top Predictions for 2024 blog series.
A New Breed of Cost Optimization
So what’s this new breed of cost optimization all about? Short answer: giving engineering teams more power to optimize.
Over the years, traditional FinOps tools created platforms that provide general information about the resources being used, the cost associated to them, and recommendations for improvements. While these tools can start bridging the gap between teams, the manual optimization required for analysis and remediation may place limits on their consistent adoption cross-functionally.
In comparison, this new wave of solutions is designed to be implemented closer to the engineering teams. While they are focused more on the engineering side with resource optimization, they also optimize costs along the way and provide those metrics for the FinOps team to use. This shift from FinOps to engineering is fueling this new era of engineering optimization.
Early Signals of this Engineering Optimization Evolution
We’re seeing some early signals of this evolution with solutions taking advantage of this in different ways. On this journey to engineering automated optimization, there are solutions that are providing observability into engineering resources and optimization recommendations of those resources (examples include BlueSky, Revefi, Infracost, amongst others). While these provide value closer to engineering teams, there are solutions going farther to provide tailored insights into the performance of resources and find automatic optimization opportunities to reduce the manual effort. Due to the knowledge required to provide this engineering optimization lens, emerging solutions are typically narrow in their expertise and their offerings are dependent on the resources ability to be optimized. Each cloud provider and infrastructure component has a different offering structure, with some being easier to identify optimization opportunities than others. As a result, vendors have flocked to those easy win areas and the more difficult resources to optimize have fewer offerings. A prime example of this is AWS Cloud Computing- due to AWS’ flexible purchase options, many vendors have created offerings around optimizing AWS Savings Plans and Reserved Instances. Azure and GCP have fewer of these offerings, as they naturally have a different cloud computing structure.
Although cloud computing was the example above, solutions are spanning across many areas and our team expects new automated optimization approaches to continue evolving as this engineering optimization trend takes off. In the meantime, below is a list of signals we are currently tracking in the automated optimization space:
-
Cloud Computing Optimization (AWS)
Hybrid and multi-cloud environments can make it challenging to ensure visibility over cloud resources, as each cloud service provider differs in their cloud computing offerings. They also differ in their ease for optimization and therefore, while AWS is represented in the automated optimization market, Azure and GCP are more commonly seen in visibility reporting solutions. This is mainly due to AWS’ flexible purchase options, allowing solutions in this space to automatically right-size RIs and other instance plans and provide the highest level of discounts.
AWS Computing Optimization: ProsperOps, Zesty, Exostellar
-
Infrastructure Optimization
At the infrastructure layer, there are a variety of different opportunities for ensuring performance and cost optimization. Kubernetes optimization solutions have become popular for analyzing cluster configurations and automatically scaling resources to optimize performance, and therefore cost.
Kubernetes Optimization: ScaleOps, Sedai, Cast AI
-
Data Optimization
Enterprises continuously leverage more and more data, resulting in increased data costs for management, transport, governance, and storage. To solve this, solutions are providing cloud optimization opportunities for cloud engineering resources. Our team is tracking approaches across data infrastructure, including compression and rehydration for data management and GPU utilization optimization for AI/ML infrastructure. With the boom of GenAI, we are seeing signs in the AI/ML infrastructure market of solutions tailoring their offerings to help with the extensive resources required for deploying and managing generative AI.
AI/ML Infrastructure: Run.ai , Deci, OctoAI
Data Management: Granica, Ultihash
*Note: Solutions listed above are sample solutions in this space and are not meant to be representative of the full market.
As seen above, the signals arising in the market are grouped by optimization function. Since these solutions are narrow in their focus areas, we recommend determining your use cases to begin taking advantage of optimization solutions.
The Future of Cost Optimization
The future is always tricky to make bets on but signals in this market are leading us to certain outlooks. While we focused on the investment signals of best of breed engineering optimization solutions, our Innovation team predicts this approach to cost management will extend beyond these solution sets. In fact, we are already starting to see traditional FinOps visibility reporting platforms extending capabilities to automated optimization solutions, either through partnerships or proprietary development. Additionally, there are signs to show that solutions are expanding their optimization scope into other areas where the use cases make sense. This further validates that there is something special here and that with shifting more towards an engineering optimization approach, FinOps teams and developers can work together to keep cloud costs in order. If you’re interested in learning more, please don’t hesitate to contact us at Innovation@trace.com.
Kiersten Putnam is a Senior Innovation Researcher at Trace3. She is passionate about new innovative approaches that challenge traditional processes across the enterprise. As a member of the Innovation Team, she delivers research content on emerging trends and solutions across enterprise cloud, security, data, and infrastructure. When she's not researching, she is either exploring the surrounding areas of Denver, Colorado where she lives, or planning her next trip abroad.