Leveraging Interpretability in the Transformer to Automate the Proactive Scaling of Cloud Resources
This work addresses the problem of optimizing resource scaling for cloud microservices to ensure QoS and reduce costs, presenting an incremental improvement by combining existing interpretable models.
The paper tackles the challenge of accurately provisioning cloud microservices to meet latency SLAs by developing a model that predicts end-to-end latency using a Temporal Fusion Transformer and adjusts resource allocation via Kernel Ridge Regression based on feature importance.
Modern web services adopt cloud-native principles to leverage the advantages of microservices. To consistently guarantee high Quality of Service (QoS) according to Service Level Agreements (SLAs), ensure satisfactory user experiences, and minimize operational costs, each microservice must be provisioned with the right amount of resources. However, accurately provisioning microservices with adequate resources is complex and depends on many factors, including workload intensity and the complex interconnections between microservices. To address this challenge, we develop a model that captures the relationship between an end-to-end latency, requests at the front-end level, and resource utilization. We then use the developed model to predict the end-to-end latency. Our solution leverages the Temporal Fusion Transformer (TFT), an attention-based architecture equipped with interpretability features. When the prediction results indicate SLA non-compliance, we use the feature importance provided by the TFT as covariates in Kernel Ridge Regression (KRR), with the response variable being the desired latency, to learn the parameters associated with the feature importance. These learned parameters reflect the adjustments required to the features to ensure SLA compliance. We demonstrate the merit of our approach with a microservice-based application and provide a roadmap to deployment.