Harness today extended its platform for optimizing the consumption of cloud infrastructure resources to provide an ability to enforce governance policies as code.
In addition, the latest version of the Harness Cloud Cost Management platform now provides an ability to leverage machine learning algorithms to more accurately forecast spending at the instance level on the Amazon Elastic Compute Cloud (EC2). It then automates the purchasing and management of Reserved Instances (RIs) and Savings Plans to better align cloud infrastructure spending patterns over time.
Finally, Harness also includes a tool that makes it simpler to automatically scale instances of Kubernetes cluster running on the Amazon Elastic Kubernetes Service (EKS). It can also incorporate less expensive Spot instances of cloud infrastructure to provide that additional capacity when needed.
Ravi Yadalam, director of product management for Harness, said collectively these capabilities will make it simpler for FinOps teams to maximize usage of cloud resources at a time when the number of applications being deployed in cloud computing environments continues to accelerate.
Harness already makes available a generative artificial intelligence (AI) assistant, dubbed AI Development Assistant or AIDA, that can be used to create governance rules that can now be automatically applied as code using YAML files to, for example, upgrading to less expensive or faster storage service as needed. That Cloud Asset Governance capability can be applied across Amazon Web Services (AWS), Microsoft Azure or the Google Cloud Platform (GCP).
With the rise of FinOps as a methodology for applying best practices to manage cloud infrastructure, organizations are trying to accomplish two goals. Historically, many developers have provisioned cloud resources directly using a corporate credit card. However, developers tend to over-provision cloud infrastructure resources to ensure application availability. Not being woken up in the middle of the night because there is a resource issue is more important to a developer than the cost of a cloud service. Unfortunately, all the over-provisioning also leads to a lot of wasted cloud spending.
At the same time, organizations also want to deploy more applications than ever in the cloud. Thanks to the rise of AI the cost of building and deploying applications is rapidly declining but organizations will still need to make sure cloud infrastructure is being used optimally otherwise the cost of deploying all that software is likely to become prohibitive.
Compare Cloud Cost
Fortunately, the FinOps Foundation is laying the groundwork for a specification that will make it simpler to initially compare costs across cloud services, before eventually using that normalized data to apply AI to optimally deploy workloads across a hybrid cloud computing environment.
Itโs not clear just yet how much AI will be relied on to optimize the usage of cloud infrastructure. However, AI assistants will play a larger role in managing IT infrastructure, noted Yadalam. In the short term, however, IT administrators working closely with their colleagues in the finance department are being increasingly tasked with making sure cloud costs donโt spiral out of control.