Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Chronicles of Jockeying in Queuing Systems

Anthony Kiggundu, Bin Han, Dennis Krummacker, Hans D. Schotten

TL;DR

This work surveys jockeying, the switching of jobs between queues, in latency-sensitive networks such as MEC and 5G/Beyond. It categorizes modeling approaches into stochastic, analytic, and behavioral families, analyzes their applicability under next-generation architectures, and argues that classical models are increasingly inadequate due to slice heterogeneity, partial state information, and migration costs. The authors propose design principles—hybrid centralized/decentralized architectures, value-of-information driven dissemination, and robust, transport-aware decision making—and outline open challenges like state dissemination, migration overhead, and security. The study emphasizes practical validation on MEC/NTN testbeds and advocates integrating jockeying concepts with network slicing and NFV/SDN to achieve SLA-compliant resource orchestration. Overall, the paper maps the theoretical landscape, identifies gaps for 5G+/6G deployments, and provides concrete directions for robust, scalable impatience-aware system design.

Abstract

Emerging trends in communication systems, such as network softwarization, functional disaggregation, and multi-access edge computing (MEC), are reshaping both the infrastructural landscape and the application ecosystem. These transformations introduce new challenges for packet transmission, task offloading, and resource allocation under stringent service-level requirements. A key factor in this context is queue impatience, where waiting entities alter their behavior in response to delay. While balking and reneging have been widely studied, this survey focuses on the less explored but operationally significant phenomenon of jockeying, i.e. the switching of jobs or users between queues. Although a substantial body of literature models jockeying behavior, the diversity of approaches raises questions about their practical applicability in dynamic, distributed environments such as 5G and Beyond. This chronicle reviews and classifies these studies with respect to their methodologies, modeling assumptions, and use cases, with particular emphasis on communication systems and MEC scenarios. We argue that forthcoming architectural transformations in next-generation networks will render many existing jockeying models inapplicable. By highlighting emerging paradigms such as MEC, network slicing, and network function virtualization, we identify open challenges, including state dissemination, migration cost, and stability, that undermine classical assumptions. We further outline design principles and research directions, emphasizing hybrid architectures and decentralized decision making as foundations for re-conceptualizing impatience in next-generation communication systems.

Chronicles of Jockeying in Queuing Systems

TL;DR

This work surveys jockeying, the switching of jobs between queues, in latency-sensitive networks such as MEC and 5G/Beyond. It categorizes modeling approaches into stochastic, analytic, and behavioral families, analyzes their applicability under next-generation architectures, and argues that classical models are increasingly inadequate due to slice heterogeneity, partial state information, and migration costs. The authors propose design principles—hybrid centralized/decentralized architectures, value-of-information driven dissemination, and robust, transport-aware decision making—and outline open challenges like state dissemination, migration overhead, and security. The study emphasizes practical validation on MEC/NTN testbeds and advocates integrating jockeying concepts with network slicing and NFV/SDN to achieve SLA-compliant resource orchestration. Overall, the paper maps the theoretical landscape, identifies gaps for 5G+/6G deployments, and provides concrete directions for robust, scalable impatience-aware system design.

Abstract

Emerging trends in communication systems, such as network softwarization, functional disaggregation, and multi-access edge computing (MEC), are reshaping both the infrastructural landscape and the application ecosystem. These transformations introduce new challenges for packet transmission, task offloading, and resource allocation under stringent service-level requirements. A key factor in this context is queue impatience, where waiting entities alter their behavior in response to delay. While balking and reneging have been widely studied, this survey focuses on the less explored but operationally significant phenomenon of jockeying, i.e. the switching of jobs or users between queues. Although a substantial body of literature models jockeying behavior, the diversity of approaches raises questions about their practical applicability in dynamic, distributed environments such as 5G and Beyond. This chronicle reviews and classifies these studies with respect to their methodologies, modeling assumptions, and use cases, with particular emphasis on communication systems and MEC scenarios. We argue that forthcoming architectural transformations in next-generation networks will render many existing jockeying models inapplicable. By highlighting emerging paradigms such as MEC, network slicing, and network function virtualization, we identify open challenges, including state dissemination, migration cost, and stability, that undermine classical assumptions. We further outline design principles and research directions, emphasizing hybrid architectures and decentralized decision making as foundations for re-conceptualizing impatience in next-generation communication systems.
Paper Structure (30 sections, 15 equations, 2 figures, 2 tables)

This paper contains 30 sections, 15 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Triggers and modeling variants
  3. Threshold-based switching:
  4. Cost and expected delay criteria:
  5. Queue typology and baseline assumptions
  6. Why classical assumptions may fail in MEC or 5G
  7. Implications and objectives of this survey
  8. Techniques for Modeling Jockeying in Queues
  9. Stochastic Modeling
  10. Analytic Modeling
  11. Behavioral Modeling
  12. Stochastic Models
  13. Statistical Models
  14. Limitations of Statistical Models in Next Generation Networks: Statistical modeling of jockeying in queues faces several practical limits in next-generation networks. For example, in such environments telemetry is often partial, delayed or censored, producing biased and high-variance estimates. Also, the system is very dynamic (mobility, auto-scaling, flash crowds) causing rapid concept drifts. New architectural abstractions like slice or tenant heterogeneity invalidates pooled models. And rare but useful tail events like SLA breaches, are underrepresented by standard loss functions. closed-loop effects (predictions changing behaviour) and multi-agent interactions break offline validity; edge compute and latency constraints limit model complexity and freshness; uncertainty quantification is frequently absent or miscalibrated; telemetry can be adversarially manipulated; and ground-truth labels for “beneficial” jockeying are ambiguous—altogether making naïve statistical predictors fragile unless paired with censoring-aware estimators, adaptive retraining, causal or closed-loop evaluation, and robust telemetry authentication.
  15. Nash Equilibrium based Models
  16. ...and 15 more sections

Figures (2)

  • Figure 1: Jockeying strategies: In the left-most (A) of the illustration is the Maitre d'Hotel queueing strategy where customers waited in a single line and got served when one of the available stations was empty. In the middle (B) is the Tellers' Window strategy where customers joined and waited in the shorter of the two queues and no switching lines was permitted thereafter. In the right-most (C) was the Tellers' Window with Jockeying, a behaviour where despite a new customer having joined the shorter of the two queues, switching to an alternative one was permitted later given a deviation in the sizes by one.
  • Figure 2: Illustrating the impatient customer that had the option to either process a task on a MEC server or locally depending on the latency requirements of the underlying application