SageSched: Intelligent LLM Request Scheduler with Workload Prediction — QoS-aware dual-queue scheduling for black-box LLM APIs (OpenAI/Azure/Doubao/Gemini)
-
Updated
May 18, 2026 - Python
SageSched: Intelligent LLM Request Scheduler with Workload Prediction — QoS-aware dual-queue scheduling for black-box LLM APIs (OpenAI/Azure/Doubao/Gemini)
K-AGRUED (Journal of Grid Computing, 2022): proactive container autoscaling for web applications in Kubernetes, using an attention-based GRU encoder-decoder that forecasts load several steps ahead.
Research paper and technical notes on the microservice ecosystem
ML-driven cloud VM scheduler using CPU workload forecasting to optimize resource utilization and reduce energy consumption.
Experimental pipeline for a simulation-based comparison of ML-driven auto-scaling policies (LSTM and Random Forest) against Kubernetes HPA, using the Alibaba Cluster Trace 2018 dataset. Published at UTP Student and Doctoral Scientific Session 2026.
Self-adaptive data layout for distribute joins
Budget-aware warm-container allocation for serverless workloads: the WRAP-GREEDY knapsack allocator vs. baseline policies, benchmarked by trace-driven simulation on the Huawei/Joosen public serverless trace.
Adaptive task scheduling in Apache Spark using ML-based workload prediction (Big Data Systems lab project).
To associate your repository with the workload-prediction topic, visit your repo's landing page and select "manage topics."