The Multiserver-Job Stochastic Recurrence Equation for Cloud Computing Performance Evaluation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915759093972992 |
|---|---|
| author | Baccelli, Francois Olliaro, Diletta Marsan, Marco Ajmone Marin, Andrea |
| author_facet | Baccelli, Francois Olliaro, Diletta Marsan, Marco Ajmone Marin, Andrea |
| contents | We study the Multiserver-Job Queuing Model (MJQM) with general independent arrivals and service times under FCFS scheduling, using stochastic recurrence equations (SREs) and ergodic theory. We prove the monotonicity and separability properties of the MJQM SRE, enabling the application of the monotone-separable extension of Loynes' theorem and the formal definition of the MJQM stability condition. Based on these results, we introduce and implement two algorithms: one for drawing sub-perfect samples (SPS) of the system's workload and the second one to estimate the system's stability condition given the statistics of the jobs' input stream. The SPS algorithm allows for a massive GPU parallelization, greatly improving the efficiency of performance metrics evaluation. We also show that this approach extends to more complex systems, including MJQMs with typed resources. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20653 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | The Multiserver-Job Stochastic Recurrence Equation for Cloud Computing Performance Evaluation Baccelli, Francois Olliaro, Diletta Marsan, Marco Ajmone Marin, Andrea Performance We study the Multiserver-Job Queuing Model (MJQM) with general independent arrivals and service times under FCFS scheduling, using stochastic recurrence equations (SREs) and ergodic theory. We prove the monotonicity and separability properties of the MJQM SRE, enabling the application of the monotone-separable extension of Loynes' theorem and the formal definition of the MJQM stability condition. Based on these results, we introduce and implement two algorithms: one for drawing sub-perfect samples (SPS) of the system's workload and the second one to estimate the system's stability condition given the statistics of the jobs' input stream. The SPS algorithm allows for a massive GPU parallelization, greatly improving the efficiency of performance metrics evaluation. We also show that this approach extends to more complex systems, including MJQMs with typed resources. |
| title | The Multiserver-Job Stochastic Recurrence Equation for Cloud Computing Performance Evaluation |
| topic | Performance |
| url | https://arxiv.org/abs/2601.20653 |