Operations | Monitoring | ITSM | DevOps | Cloud

On Oct. 20, the Internet woke up and seemingly chose violence. For more than 12 hours, Amazon Web Services (AWS) went down. From banking platforms to hospital communications to mobile ordering apps, digital services came to a screeching halt. The cause? Two programs are trying to write a DNS entry simultaneously, failing, and leaving the entry blank. Thus began the incredibly costly failure cascade.

Before 2022, Alex Hanna worked on Google’s Ethical AI team. Today, she’s the director of research at the Distributed AI Research Institute, a transition sparked by Google’s handling of a paper exposing AI’s growing environmental footprint. So, how bad is it, really? That depends on who you ask. Take Jesse Dodge, a senior research analyst at the Allen Institute for AI. Jesse told NPR that a single ChatGPT query can use as much electricity as keeping a light bulb on for 20 minutes.

In a webinar hosted by CloudZero on Oct. 30, 2025, Larry Advey, Director of Cloud Platform and FinOps and a respected voice in the FinOps community, joined Umesh Rao to deliver a practical session on cloud cost allocation. The session, titled Three Allocation Mistakes Most FinOps Teams Make, unpacked hard-earned lessons and offered a guided tour of CloudZero’s new Dimension Studio.

Imagine you have an API service composed of multiple microservices. Traffic fluctuates — sometimes light, sometimes spiking. Without Fargate, you’d have to manage EC2 instances, autoscaling, patching, and more. With Fargate, you define each microservice as a task, setting the CPU/memory, container image, network rules, and AWS schedules, and then run them as needed. The result: faster deployment, lower ops overhead, and smooth scaling.

If Meta’s (now halted) nine-figure AI talent poaching scheme was any indication, the AI talent market is pretty frothy. The number of AI-related job postings has roughly tripled since 2019, and the average salary has more than doubled (Bain). The race is on for companies to find the fastest, most sustainable routes to AI-driven business value; all companies, but especially software companies, are hotly pursuing racers. But despite what Zuckerberg & Co.

Before 2011, cloud infrastructure was still new. AWS had launched EC2 and S3 in 2006. But to deploy applications, engineers had to manually spin up servers, configure storage, and set up networking — all by hand or with custom scripts. There were early configuration management tools, such as Chef and Puppet, but those didn’t offer full cloud orchestration. Then in 2011, AWS launched AWS CloudFormation as the first major orchestration tool.

Every finance team knows this pain. The cloud bill continues to grow, but the numbers don’t quite add up. Engineering swears their usage reports are accurate. But when you ask, “Which product or customer drove last month’s 18% cost increase?” things go quiet. That silence usually means one thing. Inaccurate cost allocation. Native cloud cost allocation tools from AWS, Azure, and Google Cloud can help. However, they often stop at the average this or average that layer.

SaaS, AI, and technology spending today looks like a more intense version of how it was a decade ago when everyone first migrated to the cloud. The mentality was and, at some companies, still is to build build build and worry about controlling expenditures and optimizing costs later. We’re seeing astronomical amounts of money being raised by brand-new AI businesses that barely even existed a couple of years ago. More than $145 billion has been raised for U.S.

One of the most challenging aspects of running a SaaS business is dealing with highly variable costs. Every service you provide needs resources, and the demand for those resources fluctuates. Just when you think you’ve got this quarter’s budget nailed down, an engineer makes a restructuring decision or a new feature launch exceeds expectations (or falls flat). Suddenly, your bill looks totally different than what you were expecting.

AWS did not intend to build Amazon EKS; it simply had to. Kubernetes adoption beamed light years ahead of AWS’s own managed container orchestration service. This forced AWS to develop a managed service to accommodate customers who wanted to use upstream Kubernetes but did not want to manage it themselves. As soon as AWS got around to it, it knocked the Kubernetes-based container management service out of the park. Not only is Amazon EKS simpler than Kubernetes, but EKS pricing may also be worth it.