Published: 19 August 2024
Summary
Developers using cloud-native and generative AI technologies need to self-serve, provision, manage and scale infrastructure on demand. Platform engineering managers can engage the cool vendors in this research to simplify infrastructure-related workflows for software engineering teams.
Included in Full Research
Overview
Key Findings
-
Optimizing compute costs, performance and sustainability becomes difficult as product teams scale application delivery across multiple environments and multiple clouds.
-
Inconsistent infrastructure automation practices in cloud environments can create misconfiguration-related security risks and cause cloud costs to spiral out of control.
-
Developers continue to seek easier methods of self-provisioning and managing development environments and DevOps infrastructure without having to manually code intricate configuration details.
-
Product teams building generative AI (GenAI) applications need access to scalable infrastructure with graphics processing unit (GPU) acceleration for model training, fine-tuning and inferencing.
Recommendations
Platform engineering managers should:
Clients can log in to view the entire
document.