Aasish Kumar Sharma

NHR@Göttingen, Göttingen

Research Topic

Optimization of workload mapping and scheduling in heterogeneous High Performance Computing (HPC) systems

 

Supervisor

  • Prof. Dr. Julian M. Kunkel
  • Prof. Dr. Ramin Yahyapour
  • Prof. Dr. Fabian Sinz

 

Thesis abstract

Optimization of Workload Mapping and Scheduling in Heterogeneous HPC System Landscapes

This thesis investigates efficient workload mapping and scheduling in heterogeneous HPC systems. It develops a unified optimization framework and systematically evaluates exact optimization, heuristics, metaheuristics, quantum-inspired methods, and graph neural network-based reinforcement learning. Through reproducible benchmarks across increasingly large and heterogeneous workflow instances, the work identifies the capabilities and scalability limits of different scheduling approaches and develops learning-based methods for scalable, constraint-aware scheduling when exact optimization becomes computationally impractical.
Name of Supervisor: Prof. Dr. Julian M. Kunkel
 

Secondment

NHR@ZIB


Publications

  • Aasish Kumar Sharma, Julian M. Kunkel, et al. GrapheonRL: A GNN and Reinforcement Learning Framework for Constraint and Data-Aware Workflow Mapping and Scheduling in Heterogeneous HPC Systems. IEEE COMPSAC 2025.
  • Aasish Kumar Sharma, Christian Boehme, Philipp Gelß, Ramin Yahyapour, Julian M. Kunkel. Workflow-Driven Modeling for the Compute Continuum: An Optimization Approach to Automated System and Workload Scheduling. IEEE COMPSAC 2025.
       
  • Aasish Kumar Sharma, S. P. Pandey, Julian M. Kunkel. Performance Analysis of Convolutional Neural Network by Applying Unconstrained Binary Quadratic Programming. IEEE COMPSAC 2025.
      
  • Aasish Kumar Sharma, Christian Boehme, Julian M. Kunkel, et al. An Empirical Evaluation of Quantum-Inspired QUBO Methods for Heterogeneous HPC Workflow Mapping and Scheduling. ISC High Performance 2026.
      
  • L. van Niekerk, D. Kumar, Aasish Kumar Sharma, et al. A Comparison of HPC-Based Quantum Computing Simulators Using Quantum Volume. INFORMATIK 2024.
       
  • Aasish Kumar Sharma, Julian M. Kunkel. A Review of Tools and Techniques for Optimization of Workload Mapping and Scheduling in Heterogeneous HPC Systems. arXiv:2505.11244, 2025.
       

  • Aasish Kumar Sharma et al. Evaluating Large Language Models for Workload Scheduling in HPC Systems. arXiv:2511.11612, 2025.

     

Conference presentations

  • SC26, Chicago, USA: short paper: "Repository-Scale Performance Characterization of the IO500 Benchmark”, presentation at the 17th International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS), held in conjunction with SC26.
       
  • ISC High Performance 2026, Hamburg, Germany: Research paper presentation, An Empirical Evaluation of Quantum-Inspired QUBO Methods for Heterogeneous HPC Workflow Mapping and Scheduling.
       
  •  ISC High Performance 2026, Hamburg, Germany: Co-organizer/instructor, tutorial Quantum Computing for HPC Experts.
       
  • IEEE COMPSAC 2025: Presentation of research on GrapheonRL and heterogeneous HPC workflow scheduling.
  • SC25, St. Louis, USA: Doctoral Showcase poster presentation.
       

  • ISC High Performance 2025, Hamburg, Germany: Student Research Poster presentation.

     

Awards / Selected Recognition

  • SC25 Doctoral Showcase, selected doctoral research poster.
  • NHR Graduate School scholarship/fellowship, if formally applicable to your status.