A multi-agent architecture for scheduling of high performance services in a GPU cluster

  • Joel Antonio Trejo-Sánchez CONACYT-Centro de Investigación en Matemáticas
  • José Luis López-Martínez Universidad Autónoma de Yucatán- Facultad de Matemáticas
  • José Octavio Gutiérrez García ITAM
  • Julio César Ramírez-Pacheco Universidad del Caribe- Departamento en Ciencias Básicas e Ingenierías
  • Daniel Fajardo-Delgado Instituto Tecnológico de Cd Guzmán
Keywords: High performance computing, scheduling


Nowadays, clusters containing multiple GPU nodes are widely used to execute high-performance computing applications. Diverse disciplines use these clusters to improve the performance of several services that consume high computational resources. The challenge of executing high-performance computing applications becomes harder when the applications are executed concurrently and each one of them may demand multiple GPU nodes for different periods of time. To tackle this challenge, we propose a multi-agent architecture for scheduling multiple services   in a heterogeneous GPU cluster. We provide simulation results of our agent-based system utilizing three commonly used scheduling heuristics for several configuration settings.

How to Cite
Trejo-Sánchez, J. A., López-Martínez, J. L., Gutiérrez García, J. O., Ramírez-Pacheco, J. C., & Fajardo-Delgado, D. (2018). A multi-agent architecture for scheduling of high performance services in a GPU cluster. International Journal of Combinatorial Optimization Problems and Informatics, 9(1), 12-22. Retrieved from https://ijcopi.org/index.php/ojs/article/view/75