AI
All inefficiencies
12 inefficiencies
| Cloud | Reference | Inefficiency | Service | Category |
|---|---|---|---|---|
 | CER-0333 | Idle SageMaker Notebook Instances Left Running ContinuouslyAWS SageMaker / AI | AWS SageMaker | AI |
| CER-0330 | Orphaned MLflow Training Artifacts and Model Checkpoints in Object StorageAWS S3 / AI | AWS S3 | AI |
 | CER-0267 | Using High-Cost Models for Low-Complexity TasksGCP Vertex AI / AI | GCP Vertex AI | AI |
| CER-0276 | Using High-Cost Bedrock Models for Low-Complexity TasksAWS Bedrock / AI | AWS Bedrock | AI |
 | CER-0238 | Always-On PTUs for Seasonal or Cyclical Azure OpenAI WorkloadsAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0248 | Non-Production Azure OpenAI Deployments Using PTUs Instead of PAYGAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0266 | Underutilized PTU Quota for Azure OpenAI DeploymentsAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0256 | Suboptimal Bedrock Inference Profile ModelAWS Bedrock / AI | AWS Bedrock | AI |
| CER-0246 | Missing Reserved PTUs for Steady-State Azure OpenAI WorkloadsAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0254 | Suboptimal Azure OpenAI Model TypeAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0277 | Using High-Cost Models for Low-Complexity TasksAzure Cognitive Services / AI | Azure Cognitive Services | AI |
| CER-0252 | Provisioned Throughput OpenAI Deployment in Non-Production EnvironmentsAzure Cognitive Services / AI | Azure Cognitive Services | AI |