Workshop Program (Accepted Papers)

• Robert Chab, Fei Li and Sanjeev Setia. Prediction-Augmented Scheduling for Minimizing GPU Fragmentation
• Ketan Maheshwari, Dalibor Klusáček, Frédéric Suter and Wesley Brewer. Workload Traces from the Frontier Exascale Supercomputer
• Kyrian Adimora and Hongyang Sun. MARS-HPC: A Multi-Objective Adaptive RL-Scheduler for Energy-Aware HPC Resource Management
• Viktória Spišaková, Dalibor Klusáček, Lukáš Hejtmánek and Radostin Stoyanov. Opportunistic Resource Reclamation in Kubernetes: From Aggressive Resizing to Flash Jobs
• Liudong Zuo, Pan Lai, Michelle Zhu and Chase Wu. Duration-Aware Bandwidth Reservation over Two Node-Disjoint Paths in High-Performance Networks
• Kanglin Xu, Zhiling Lan and Michael Papka. SafeWall: A Risk-Aware Job Walltime Recommendation for HPC Scheduling
• Sana Taghipour Anvari and David Kaeli. A Granularity Characterization of Task Scheduling Effectiveness
• Leszek Sliwko and Thorsten Neumann. Pareto Frontier Optimization for NLP-based Cluster Allocation Directives
• Niccolò Brembilla, Zhiling Lan and Michael E. Papka. From Petascale to Exascale: Evolving User Behavior on the Polaris and Aurora Systems
• Melanie Cornelius, Greg Cross, Shilpika Shilpika, Zhiling Lan and Michael E. Papka. Fusing System Data to Navigate Power-Saving Opportunities
• Brenton Walker and Markus Fidler. Patience vs Restraint as Self-adaptive Scheduling Strategies for Parallel Systems with Barriers
• Aaron Jezghani, Juyeop Baek, Deepa Phanish, Michael Weiner and Jeffrey Young. Sustainable On-Demand HPC through Managed Oversubscription
• Vasudha Devarakonda, Dat Nguyen and Khanh Nguyen. GOA: Memory-Safe Oversubscription for Deep Learning Training Workloads