Predictive models for computing clusters: a case study based on Ilifu

Simple item page
dc.contributor.advisorFrank, Bradley
dc.contributor.advisorRakotonirainy, Rosephine
dc.contributor.authorCloete, Jacobus Marthinus
dc.date.accessioned2026-07-22T09:50:42Z
dc.date.available2026-07-22T09:50:42Z
dc.date.issued2026
dc.date.updated2026-07-22T09:47:46Z
dc.description.abstractThis thesis investigates the feasibility of simulating job waiting times on a High-Performance Computing (HPC) cluster using a simplified discrete event simulation approach. We develop SimPE, a model that uses a Processor Equivalent (PE) metric to reduce the complexity of multi-resource job scheduling to a single dimension. The model is tested against a real Slurm Training Cluster and a Slurm Simulator for various scheduling policies including First-In-First-Out (FIFO), static and dynamic priority, and backfill scheduling. Results show that SimPE effectively models FIFO scheduling with waiting time estimates within 12% of the actual values. For static priority scheduling and dynamic priority scheduling, the model maintains similar levels of accuracy. However, for backfill scheduling, SimPE had more aggressive backfilling which led to much lower waiting time estimates. When modelling jobs with multiple resource requirements, SimPE accurately estimates when one resource dominates, but becomes too conservative when different resources are equally important. This research shows that while PE-based simulation can effectively model certain aspects of cluster behaviour, careful thought is needed when applying it to more complex scheduling scenarios. The findings help to increase our understanding of simple cluster simulation approaches and their practical limitations.
dc.identifier.apacitationCloete, J. M. (2026). <i>Predictive models for computing clusters: a case study based on Ilifu</i>. (). University of Cape Town ,Faculty of Science ,Department of Statistical Sciences. Retrieved from http://hdl.handle.net/11427/43631en_ZA
dc.identifier.chicagocitationCloete, Jacobus Marthinus. <i>"Predictive models for computing clusters: a case study based on Ilifu."</i> ., University of Cape Town ,Faculty of Science ,Department of Statistical Sciences, 2026. http://hdl.handle.net/11427/43631en_ZA
dc.identifier.citationCloete, J.M. 2026. Predictive models for computing clusters: a case study based on Ilifu. . University of Cape Town ,Faculty of Science ,Department of Statistical Sciences. http://hdl.handle.net/11427/43631en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Cloete, Jacobus Marthinus AB - This thesis investigates the feasibility of simulating job waiting times on a High-Performance Computing (HPC) cluster using a simplified discrete event simulation approach. We develop SimPE, a model that uses a Processor Equivalent (PE) metric to reduce the complexity of multi-resource job scheduling to a single dimension. The model is tested against a real Slurm Training Cluster and a Slurm Simulator for various scheduling policies including First-In-First-Out (FIFO), static and dynamic priority, and backfill scheduling. Results show that SimPE effectively models FIFO scheduling with waiting time estimates within 12% of the actual values. For static priority scheduling and dynamic priority scheduling, the model maintains similar levels of accuracy. However, for backfill scheduling, SimPE had more aggressive backfilling which led to much lower waiting time estimates. When modelling jobs with multiple resource requirements, SimPE accurately estimates when one resource dominates, but becomes too conservative when different resources are equally important. This research shows that while PE-based simulation can effectively model certain aspects of cluster behaviour, careful thought is needed when applying it to more complex scheduling scenarios. The findings help to increase our understanding of simple cluster simulation approaches and their practical limitations. DA - 2026 DB - OpenUCT DP - University of Cape Town KW - Data science LK - https://open.uct.ac.za PB - University of Cape Town PY - 2026 T1 - Predictive models for computing clusters: a case study based on Ilifu TI - Predictive models for computing clusters: a case study based on Ilifu UR - http://hdl.handle.net/11427/43631 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/43631
dc.identifier.vancouvercitationCloete JM. Predictive models for computing clusters: a case study based on Ilifu. []. University of Cape Town ,Faculty of Science ,Department of Statistical Sciences, 2026 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/43631en_ZA
dc.language.isoen
dc.language.rfc3066Eng
dc.publisher.departmentDepartment of Statistical Sciences
dc.publisher.facultyFaculty of Science
dc.publisher.institutionUniversity of Cape Town
dc.subjectData science
dc.titlePredictive models for computing clusters: a case study based on Ilifu
dc.typeThesis / Dissertation
dc.type.qualificationlevelMasters
dc.type.qualificationlevelMSc
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
thesis_sci_2026_cloete jacobus marthinus.pdf
Size:
3.05 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.72 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Masters