Deterministic Task Offloading and Resource Allocation in the IoT-Edge-Cloud Continuum

Baldomero Coll-Perales; Javier Gozalvez; Keyvan Aghababaiyan

arxiv: 2604.16155 · v1 · submitted 2026-04-17 · 💻 cs.NI

Deterministic Task Offloading and Resource Allocation in the IoT-Edge-Cloud Continuum

Keyvan Aghababaiyan , Baldomero Coll-Perales , Javier Gozalvez This is my paper

Pith reviewed 2026-05-10 07:18 UTC · model grok-4.3

classification 💻 cs.NI

keywords task offloadingresource allocationIoT-edge-cloud continuumdeterministic servicesflexible latenciescommunication and computing resourcestime-sensitive applicationsnetwork performance

0 comments

The pith

A deterministic scheme for task offloading in IoT-edge-cloud systems meets more deadlines by allowing flexible task latencies instead of minimizing each one.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper proposes a scheme for jointly managing communication and computing resources across IoT devices, edge servers, and the cloud. Rather than optimizing for the shortest possible execution time on every task, it ensures tasks finish before their deadlines even if that means using somewhat longer but still acceptable times for some of them. The approach uses these flexible latencies to free up resources and accommodate more tasks when overall capacity is limited. A sympathetic reader would care because future networks must deliver reliable, time-bounded performance for critical applications while scaling efficiently under real resource constraints.

Core claim

The paper presents a deterministic task offloading and resource allocation scheme for the IoT-edge-cloud continuum that prioritizes task completion before deadlines over minimizing the latency in the execution of individual tasks. The scheme leverages flexible latencies across tasks to support a higher number of tasks through a more efficient management of computing and communication resources that better adapts to scenarios with constrained resources.

What carries the argument

The deterministic task offloading and resource allocation scheme that jointly manages communication and computing resources by trading flexible latencies to meet deadlines.

If this is right

The system supports a higher number of tasks in resource-constrained environments while still meeting deadlines.
It enables deterministic service levels for critical and time-sensitive vertical applications.
Opportunistic offloading across the continuum enhances overall network performance and allows scaling with demand.
Joint management of communication and computing resources becomes more adaptable to varying loads.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same deadline-first logic could be tested in other distributed systems such as fog computing or vehicular networks where task timing tolerances vary.
Real-time monitoring of actual task flexibility might allow the scheme to adapt dynamically without prior knowledge of all latencies.
Energy or power consumption trade-offs when stretching some tasks could be measured as a next-step metric in constrained devices.

Load-bearing premise

Tasks possess adjustable flexible latencies that can be traded off without violating their individual service requirements, and the IoT-edge-cloud continuum model accurately represents real-world network conditions and resource constraints.

What would settle it

A simulation or testbed run in which minimizing every task's latency supports at least as many tasks meeting all deadlines as the flexible-latency scheme does, or in which the flexible approach causes deadline violations under realistic traffic and resource traces.

read the original abstract

Future cellular networks will sustainably integrate computing, intelligence and services within a network of networks ecosystem that includes IoT devices and subnetworks for local communications and distributed processing. This integration creates an IoT-edge-cloud continuum that enables opportunistic task offloading across the continuum, enhancing network performance, reducing response times and allowing a flexible resource allocation that can facilitate the system to scale according to demand. Future networks should also natively support deterministic service levels for critical and time-sensitive vertical applications. In this paper, we propose a deterministic task offloading and resource allocation scheme for the joint management of communication and computing resources in the IoT-edge-cloud continuum. The proposed scheme prioritizes task completion before deadlines over minimizing the latency in the execution of individual tasks. The scheme leverages flexible latencies across tasks to support a higher number of tasks through a more efficient management of computing and communication resources that better adapts to scenarios with constrained resources.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper proposes prioritizing deadlines over min latency in IoT-edge-cloud offloading by trading flexible per-task times, but supplies no model, bounds, or results to show the trade-off works.

read the letter

The main takeaway is that this paper puts forward a deterministic task offloading and resource allocation scheme for the IoT-edge-cloud continuum, where the focus is on ensuring tasks meet their deadlines by leveraging flexible latencies rather than strictly minimizing execution times for each one. This approach aims to handle more tasks in constrained environments, which is relevant for time-sensitive applications in future networks. However, the description stays at a high level without backing up the performance claims. On the positive side, the work identifies a useful distinction in how to manage resources when dealing with deterministic services. By prioritizing deadline adherence over individual latency reduction, it opens up possibilities for better overall system efficiency in scenarios where some tasks can tolerate a bit more delay as long as they finish on time. The framing around the IoT-edge-cloud integration and the need for joint communication and computing management is clear and timely, given the push toward 6G and vertical applications that require reliable timing. The main weakness is the absence of any concrete model or validation for the flexible latency idea. The central claim depends on tasks having adjustable completion windows that the allocator can exploit without causing new deadline misses, but the paper does not define what those windows look like or provide any bounds or proofs to ensure safety. There are no algorithm details, mathematical formulations, simulation setups, or comparisons to existing offloading techniques mentioned. This leaves the assertion that the scheme supports a higher number of tasks as an untested hypothesis rather than a demonstrated result. The stress-test note captures this accurately: without an explicit slack interval and a guarantee that the lower bound is respected, the efficiency gain could be illusory. If the full manuscript includes detailed derivations, pseudocode for the allocation, and numerical results showing gains over baselines, that would change the picture substantially. As presented in the abstract, though, the contribution feels preliminary. This kind of paper would appeal to researchers in computer networking, particularly those focused on edge computing, resource allocation, and support for deterministic networking in IoT settings. A reader who wants high-level ideas or problem statements for future work could find it useful as a starting point. But for someone seeking rigorous analysis or reproducible methods, it falls short at this stage. I would not recommend sending it for peer review in its current state. The idea has potential, but it needs substantial development on the modeling and evaluation side before it is ready for serious refereeing.

Referee Report

1 major / 1 minor

Summary. The manuscript proposes a deterministic task offloading and resource allocation scheme for joint management of communication and computing resources in the IoT-edge-cloud continuum. The scheme prioritizes completing tasks before their deadlines rather than minimizing per-task latency, by exploiting flexible latencies across tasks to admit a larger number of tasks under constrained resources.

Significance. If the scheme can be shown via explicit modeling and validation to exploit per-task latency slack without violating deterministic deadlines, the result would be significant for scaling time-sensitive IoT services in integrated edge-cloud systems, offering a resource-efficient alternative to strict latency-minimization approaches.

major comments (1)

[Abstract] Abstract (and full text as presented): the central claim that the scheme 'leverages flexible latencies across tasks' to support more tasks while meeting deadlines is load-bearing but lacks any mathematical definition of the per-task slack interval [D_min, D_max], any bound on safe flexibility, or a proof that the joint offloading/resource allocator respects the lower bound without creating new violations. This directly matches the stress-test concern and prevents assessment of the headline advantage.

minor comments (1)

The abstract is concise but the manuscript should supply the algorithm pseudocode, optimization formulation, and simulation setup/results to substantiate the claims, as none appear in the provided text.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. The concern regarding the formalization of flexible latencies is well-taken and highlights an opportunity to strengthen the presentation of our core contribution. We address the point below and commit to a major revision that incorporates the requested mathematical details.

read point-by-point responses

Referee: [Abstract] Abstract (and full text as presented): the central claim that the scheme 'leverages flexible latencies across tasks' to support more tasks while meeting deadlines is load-bearing but lacks any mathematical definition of the per-task slack interval [D_min, D_max], any bound on safe flexibility, or a proof that the joint offloading/resource allocator respects the lower bound without creating new violations. This directly matches the stress-test concern and prevents assessment of the headline advantage.

Authors: We agree that an explicit mathematical treatment of per-task latency flexibility is essential for rigor. In the revised manuscript we will add a dedicated subsection in the system model that defines, for each task i, the slack interval [D_min^i, D_max^i] where D_max^i is the deterministic deadline and D_min^i is the minimum achievable latency under the selected offloading target and resource allocation (derived from task size, CPU frequency, and link rate). We will introduce a safe-flexibility bound expressed as the ratio (D_max^i - D_min^i)/D_max^i and prove that the joint allocator never schedules a task below D_min^i by construction of the feasibility check. The proof will rely on the monotonicity of the latency function with respect to allocated resources and will be validated through both analytical bounds and numerical stress tests. These additions will appear in Sections III and IV of the revision. revision: yes

Circularity Check

0 steps flagged

No circularity detected; abstract presents design proposal without equations or load-bearing derivations

full rationale

The provided abstract and description contain no mathematical derivations, equations, fitted parameters, or self-citations that could form a derivation chain. The scheme is described as a proposal that prioritizes deadlines and leverages flexible latencies as a modeling assumption, not as a result derived from prior inputs by construction. No self-definitional steps, fitted predictions, or uniqueness theorems are present. The central claim rests on an unverified modeling choice (flexible latencies), which is a correctness issue rather than circularity. The derivation chain is therefore self-contained against external benchmarks with no reductions to inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract does not introduce or rely on any explicit free parameters, axioms, or invented entities.

pith-pipeline@v0.9.0 · 5462 in / 963 out tokens · 36632 ms · 2026-05-10T07:18:57.078135+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

[1]

European Vision for the 6G Network Ecosystem

U. Mikko, et al. “ European Vision for the 6G Network Ecosystem ”, 6G-IA Vision Working Group’ White Paper, Nov. 2024

work page 2024
[2]

Toward deterministic communications in 6G networks: state of the art, open challenges and the way forward,

G. P. Sharma et al. , "Toward deterministic communications in 6G networks: state of the art, open challenges and the way forward," IEEE Access, vol. 11, pp. 106898–106923, 2023

work page 2023
[3]

Multitask multiobjective deep reinforcement learning - based task offloading method for industrial Internet of Things,

J. Cai et al. , “Multitask multiobjective deep reinforcement learning - based task offloading method for industrial Internet of Things,” IEEE Internet Things J., vol. 10, no. 2, pp. 1848–1859, 2023

work page 2023
[4]

Joint task offloading and resource allocation for multi - access edge computing assisted by parked and moving vehicles,

W. Fan et al., “Joint task offloading and resource allocation for multi - access edge computing assisted by parked and moving vehicles,” IEEE Trans. Veh. Technol., vol. 71, no. 5, pp. 5314–5330, 2022

work page 2022
[5]

Joint task offloading and resource allocation for vehicular edge computing based on V2I and V2V modes,

W. Fan et al. , “Joint task offloading and resource allocation for vehicular edge computing based on V2I and V2V modes,” IEEE Trans. Intell. Transp. Syst., vol. 24, no. 4, pp. 4277–4292, 2023

work page 2023
[6]

Enhancing modular application placement in a hierarchical fog computing: A latency and communication cost - sensitive approach,

L. T. Oliveira et al., “Enhancing modular application placement in a hierarchical fog computing: A latency and communication cost - sensitive approach,” Comput. Commun., vol. 216, pp. 95–111, 2024

work page 2024
[7]

Intelligent task offloading for heterogeneous V2X communications,

K. Xiong et al. , “Intelligent task offloading for heterogeneous V2X communications,” IEEE Trans. Intell. Transp. Syst., vol. 22, no. 4, pp. 2226–2238, Apr. 2021

work page 2021
[8]

RJCC: Reinforcement -learning-based joint communicational-and-computational resource allocation mechanism for smart city IoT,

S. Xu et al. , “RJCC: Reinforcement -learning-based joint communicational-and-computational resource allocation mechanism for smart city IoT,” IEEE Internet Things J. , vol. 7, no. 9, pp. 8059 – 8076, 2020

work page 2020
[9]

Multi-hop cooperative task offloading for industrial IoT–edge–cloud computing environments,

Z. Hong et al., “Multi-hop cooperative task offloading for industrial IoT–edge–cloud computing environments,” IEEE Trans. Parallel Distrib. Syst., vol. 30, no. 12, pp. 2759–2774, 2019

work page 2019
[10]

In: 2025 IEEE 101st Vehicular Technology Conference (VTC2025-Spring), pp

W. Feng et al. , “Latency minimization of reverse offloading in vehicular edge computing,” IEEE Trans. Veh. Technol., vol. 71, no. 5, pp. 5343–5357, 2022. Fig. 9. Percentage of allocated tasks to computing units for various SINR values and number of tasks. Fig. 10. CDF of size (left) and workload (right) of offloaded tasks to 6G network in different link ...

work page doi:10.1109/vtc2025-spring65109.2025.11174312 2022
[11]

Refined definition of scenarios use cases and service requirements for in-X subnetworks,

6G-SHINE D2.2, “Refined definition of scenarios use cases and service requirements for in-X subnetworks,” Feb. 2023

work page 2023
[12]

Solving NP -complete Problems Using Genetic Algorithms

B. H. Arabi, "Solving NP -complete Problems Using Genetic Algorithms", in Proc. IEEE UKSim, pp. 43-48, Cambridge, UK, 2016

work page 2016
[13]

Case Study of Scaled-Up SKT 5G MEC Reference Architecture

Intel, "Case Study of Scaled-Up SKT 5G MEC Reference Architecture", White Paper, 2022

work page 2022
[14]

3GPP TS 36.211 v18.0.1 (2024), Evolved Universal Terrestrial Radio Access (E-UTRA); Physical Channels and Modulation (Release 18)

work page 2024
[15]

Heterogeneous edge offloading with incomplete information: A minority game approach ,

Hu, M., et al. "Heterogeneous edge offloading with incomplete information: A minority game approach ," IEEE Trans. Parallel Distrib. Syst. vol. 31, no. 9, pp. 2139-2154, 2020

work page 2020

[1] [1]

European Vision for the 6G Network Ecosystem

U. Mikko, et al. “ European Vision for the 6G Network Ecosystem ”, 6G-IA Vision Working Group’ White Paper, Nov. 2024

work page 2024

[2] [2]

Toward deterministic communications in 6G networks: state of the art, open challenges and the way forward,

G. P. Sharma et al. , "Toward deterministic communications in 6G networks: state of the art, open challenges and the way forward," IEEE Access, vol. 11, pp. 106898–106923, 2023

work page 2023

[3] [3]

Multitask multiobjective deep reinforcement learning - based task offloading method for industrial Internet of Things,

J. Cai et al. , “Multitask multiobjective deep reinforcement learning - based task offloading method for industrial Internet of Things,” IEEE Internet Things J., vol. 10, no. 2, pp. 1848–1859, 2023

work page 2023

[4] [4]

Joint task offloading and resource allocation for multi - access edge computing assisted by parked and moving vehicles,

W. Fan et al., “Joint task offloading and resource allocation for multi - access edge computing assisted by parked and moving vehicles,” IEEE Trans. Veh. Technol., vol. 71, no. 5, pp. 5314–5330, 2022

work page 2022

[5] [5]

Joint task offloading and resource allocation for vehicular edge computing based on V2I and V2V modes,

W. Fan et al. , “Joint task offloading and resource allocation for vehicular edge computing based on V2I and V2V modes,” IEEE Trans. Intell. Transp. Syst., vol. 24, no. 4, pp. 4277–4292, 2023

work page 2023

[6] [6]

Enhancing modular application placement in a hierarchical fog computing: A latency and communication cost - sensitive approach,

L. T. Oliveira et al., “Enhancing modular application placement in a hierarchical fog computing: A latency and communication cost - sensitive approach,” Comput. Commun., vol. 216, pp. 95–111, 2024

work page 2024

[7] [7]

Intelligent task offloading for heterogeneous V2X communications,

K. Xiong et al. , “Intelligent task offloading for heterogeneous V2X communications,” IEEE Trans. Intell. Transp. Syst., vol. 22, no. 4, pp. 2226–2238, Apr. 2021

work page 2021

[8] [8]

RJCC: Reinforcement -learning-based joint communicational-and-computational resource allocation mechanism for smart city IoT,

S. Xu et al. , “RJCC: Reinforcement -learning-based joint communicational-and-computational resource allocation mechanism for smart city IoT,” IEEE Internet Things J. , vol. 7, no. 9, pp. 8059 – 8076, 2020

work page 2020

[9] [9]

Multi-hop cooperative task offloading for industrial IoT–edge–cloud computing environments,

Z. Hong et al., “Multi-hop cooperative task offloading for industrial IoT–edge–cloud computing environments,” IEEE Trans. Parallel Distrib. Syst., vol. 30, no. 12, pp. 2759–2774, 2019

work page 2019

[10] [10]

In: 2025 IEEE 101st Vehicular Technology Conference (VTC2025-Spring), pp

W. Feng et al. , “Latency minimization of reverse offloading in vehicular edge computing,” IEEE Trans. Veh. Technol., vol. 71, no. 5, pp. 5343–5357, 2022. Fig. 9. Percentage of allocated tasks to computing units for various SINR values and number of tasks. Fig. 10. CDF of size (left) and workload (right) of offloaded tasks to 6G network in different link ...

work page doi:10.1109/vtc2025-spring65109.2025.11174312 2022

[11] [11]

Refined definition of scenarios use cases and service requirements for in-X subnetworks,

6G-SHINE D2.2, “Refined definition of scenarios use cases and service requirements for in-X subnetworks,” Feb. 2023

work page 2023

[12] [12]

Solving NP -complete Problems Using Genetic Algorithms

B. H. Arabi, "Solving NP -complete Problems Using Genetic Algorithms", in Proc. IEEE UKSim, pp. 43-48, Cambridge, UK, 2016

work page 2016

[13] [13]

Case Study of Scaled-Up SKT 5G MEC Reference Architecture

Intel, "Case Study of Scaled-Up SKT 5G MEC Reference Architecture", White Paper, 2022

work page 2022

[14] [14]

3GPP TS 36.211 v18.0.1 (2024), Evolved Universal Terrestrial Radio Access (E-UTRA); Physical Channels and Modulation (Release 18)

work page 2024

[15] [15]

Heterogeneous edge offloading with incomplete information: A minority game approach ,

Hu, M., et al. "Heterogeneous edge offloading with incomplete information: A minority game approach ," IEEE Trans. Parallel Distrib. Syst. vol. 31, no. 9, pp. 2139-2154, 2020

work page 2020