MODELING OF A TWO-PHASE QUEUING SYSTEM WITH HETEROGENEOUS SERVERS AND CUSTOMER DISCARDING
Authors:
1.
Kazan National Research Technological University
(kafedra informatiki i prikladnoy matematiki, assistent)
employee from 01.01.2023 until now
Kazan, Kazan, Russian Federation
employee from 01.01.2023 until now
Kazan, Kazan, Russian Federation
Type:
Article
Pages:
from 120
to 124
Status:
Published
Received:
31.01.2026
Accepted:
31.01.2026
Published:
31.01.2026
Language:
Russian
Keywords:
QUEUEING SYSTEMS, HETEROGENEOUS CHANNELS, PROBABILISTIC REJECTION, SIMULATION MODELING, TWO-PHASE SYSTEMS, PERFORMANCE OPTIMIZATION, PYTHON
Abstract:
Modern queuing systems often feature complex structures including heterogeneous service channels and quality control mechanisms. This study investigates the performance of two-phase queuing systems with heterogeneous channels and probabilistic request rejection. The research aims to identify effective parameter relationships to maximize throughput and minimize rejection probability. The methodological framework is based on simulation modeling in Python using NumPy and Matplotlib libraries. The developed model enables analysis of how channel heterogeneity and request filtering affect key system performance indicators. The results demonstrate that optimal distribution of channel capacities between phases enhances system resilience to peak loads. Configurations with moderate performance differences (30-50%) between channels improve key efficiency metrics by 15-25% compared to homogeneous systems. Critical system parameter values for minimal rejection probability were identified. The practical significance lies in developing recommendations for designing adaptive queuing systems in IT infrastructure, telecommunications networks, and production systems where minimizing losses and ensuring high availability are critical. The findings open new avenues for research in optimizing multiphase systems with non-stationary flows.
Modern queuing systems often feature complex structures including heterogeneous service channels and quality control mechanisms. This study investigates the performance of two-phase queuing systems with heterogeneous channels and probabilistic request rejection. The research aims to identify effective parameter relationships to maximize throughput and minimize rejection probability. The methodological framework is based on simulation modeling in Python using NumPy and Matplotlib libraries. The developed model enables analysis of how channel heterogeneity and request filtering affect key system performance indicators. The results demonstrate that optimal distribution of channel capacities between phases enhances system resilience to peak loads. Configurations with moderate performance differences (30-50%) between channels improve key efficiency metrics by 15-25% compared to homogeneous systems. Critical system parameter values for minimal rejection probability were identified. The practical significance lies in developing recommendations for designing adaptive queuing systems in IT infrastructure, telecommunications networks, and production systems where minimizing losses and ensuring high availability are critical. The findings open new avenues for research in optimizing multiphase systems with non-stationary flows.
Keywords:
QUEUEING SYSTEMS, HETEROGENEOUS CHANNELS, PROBABILISTIC REJECTION, SIMULATION MODELING, TWO-PHASE SYSTEMS, PERFORMANCE OPTIMIZATION, PYTHON
QUEUEING SYSTEMS, HETEROGENEOUS CHANNELS, PROBABILISTIC REJECTION, SIMULATION MODELING, TWO-PHASE SYSTEMS, PERFORMANCE OPTIMIZATION, PYTHON
