AWS wellarchitected documentation change
Summary
Added generative AI-specific automation recommendations including token budget enforcement, model scaling, and scheduling off-peak jobs
Security assessment
Changes focus exclusively on cost optimization through automation for generative AI workloads. No security vulnerabilities, threats, or security controls are mentioned. The added content discusses token monitoring and job scheduling purely from a cost efficiency perspective.
Diff
diff --git a/wellarchitected/latest/financial-services-industry-lens/fsicost03.md b/wellarchitected/latest/financial-services-industry-lens/fsicost03.md index 3ed08b2d6..0fce4c6ec 100644 --- a//wellarchitected/latest/financial-services-industry-lens/fsicost03.md +++ b//wellarchitected/latest/financial-services-industry-lens/fsicost03.md @@ -9 +9 @@ FSICOST03-BP01 Use automation to drive scale for Cloud Financial Management prac -Automation can drastically reduce the cost of the CFM. You can provision resources using auto scaling or using managed services, set budgets to meet, and alerts to inform users on cost utilization. +Automation can drastically reduce the cost of the CFM. You can provision resources using auto scaling or using managed services, set budgets to meet, and alerts to inform users on cost utilization. For generative AI workloads, this includes automated model scaling, token usage monitoring, and vector store optimization. @@ -15 +15,5 @@ Automation can drastically reduce the cost of the CFM. You can provision resourc -Automating operations reduces the frequency of manual tasks, improves efficiency, and benefits enterprises by delivering a consistent and reliable experience when deploying, administering, or operating workloads. You can free up human resources from manual operational tasks and use them for higher value tasks and innovations, thereby improving business outcomes. Enterprises require a proven, tested way to manage their workloads in the cloud. That solution must be secure, fast, and cost effective, with minimum risk and maximum reliability. +Automating operations reduces the frequency of manual tasks, improves efficiency, and benefits enterprises by delivering a consistent and reliable experience when deploying, administering, or operating workloads. You can free up human resources from manual operational tasks and use them for higher value tasks and innovations, thereby improving business outcomes. For example, teams can focus on improving prompt engineering or developing new AI use cases instead of managing infrastructure. Enterprises require a proven, tested way to manage their workloads in the cloud. That solution must be secure, fast, and cost effective, with minimum risk and maximum reliability. + +Automate token budget enforcement and anomaly alerts: fail‑safe requests that exceed thresholds; alert when token‑per‑call or cost‑per‑session deviates materially from baseline. + +Schedule off‑peak embedding and fine‑tuning jobs; auto‑pause development endpoints and ephemeral agents outside working hours.