Pith. sign in

REVIEW 4 major objections 4 minor 30 references

SARS: A Resource Selection Algorithm for Autonomous Driving Tasks in Heterogeneous Mobile Edge Computing

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A suitability-based resource selection algorithm with reserved processing units can push task completion rates up to 13 percent higher than classical edge-server selection in simulated autonomous driving offload.

desk verdict A clear but unreproducible RSP heuristic: the load-factor term in Eq. (8) has the wrong sign for load balancing, and the headline 13% gain comes from same-simulation parameter tuning. read the letter →

arxiv 2411.15989 v1 pith:EJOHSVMQ submitted 2024-11-24 cs.DC cs.NI

Reza Zakerian , Hadi Gholami This is my paper · ORCID
classification cs.DCcs.NI
keywords autonomousdrivingmobileedgecomputingresourceallocationreal-timetaskschedulingoffloadingcompletionratesuitabilityscoreserverheterogeneity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Offloading real-time tasks from autonomous vehicles to edge servers requires two decisions: which task to run next and which server to run it on. This paper argues that existing work has focused on the first decision and relied on crude rules, such as shortest execution time, random choice, or latest feasible finish, for the second. It proposes a suitability-based adaptive resource selection (SARS) algorithm that scores every processing unit (PU) by a weighted combination of estimated completion time, time margin before the task's deadline, and current load, then picks the highest-scoring PU that can still meet the deadline. A companion proactive on-demand resource allocation (PORA) algorithm reserves one PU per edge server near the server's average processing rate and releases it only when no ordinary PU can meet an urgent task's deadline. In simulation, the two together raise the task completion rate by up to 13% over the shortest-execution-time rule and by 3.88% over the best classical rule, across every task-selection policy tested.

What carries the argument

The load-bearing object is the suitability score $ss = t^{ct} + \alpha \cdot tm + \beta \cdot lf$, where $t^{ct}$ is the estimated completion time on a given PU, $tm$ is the time margin (deadline minus estimated completion time), and $lf$ is the load factor (the PU's current load divided by the maximum load among all PUs). SARS evaluates every PU in every edge server for each task, keeps only PUs whose predicted completion leaves a non-negative time margin, and chooses the one maximizing $ss$. PORA runs once per server: it computes the average processing rate of the server's PUs, finds the PU closest to that average, and puts it in standby; when SARS finds no feasible PU, PORA releases that reserved PU for the urgent task and later returns it to standby. This two-stage mechanism is what carries the empirical advantage.

What would settle it

Run the same 800-task simulation with fresh random draws of release times, workloads, file sizes, distances, and bandwidths, keeping $\alpha$ and $\beta$ fixed at the values from Section 5.1, and compare the TCR of SARS+PORA against the shortest-execution-time and latest-PU baselines; the central claim fails if the 13% and 3.88% margins are not reproduced. A sharper version is to replace the four task groups with a bimodal deadline distribution and check whether reserving average-rate PUs still helps.

Watch

Extended reading notes

Core claim

The central claim is that how tasks are assigned to processing units determines how many deadline-constrained tasks finish, independent of which task-selection policy is used. For every task-selection policy tested (FCFS, EDF, EDD, EFDF, CR, COVERT, and two priority-based policies), replacing the classical resource-selection rule with SARS raises the task completion rate. On its own SARS improves the best classical rule by 1.37% in the CR policy and 1.5% in EDF; when PORA is enabled, the integrated system finishes 3.88% more tasks than the latest-feasible-PU rule and 13% more than the shortest-execution-time rule in the tested scenario. The paper therefore concludes that a suitability-based RSP plus proactive reservation is a practical way to prioritize urgent tasks and increase throughput in a heterogeneous MEC architecture.

Load-bearing premise

The reported advantage depends on two tuning dials ($\alpha$ and $\beta$) being set by trial and error on the exact same simulation that is later used to measure the advantage; if those dials have to be reset whenever the tasks or servers change, the 13% gain will not carry over.

Editorial extensions

If this is right

  • Any task-selection policy (EDF, FCFS, CR, COVERT, and priority-based policies) sees a higher task completion rate when its resource-selection step is swapped to SARS, so the gain is not tied to one scheduling policy.
  • The largest reported gains require the reserved-PU mechanism: SARS with PORA beats the best classical RSP by 3.88%, while SARS alone beats it by only 1.37 to 1.5%.
  • The reserve strategy helps urgent tasks specifically, because a PU is held back precisely for tasks that no available PU can meet within their time interval.
  • SARS does not require changing task prioritization; it can be dropped into existing MEC brokers that already run EDF or other task-selection policies.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not provide a proof of optimality or a bound; the claim is empirical on one 800-task scenario, so transferability to larger fleets, multiple brokers, or different task distributions is untested.
  • The choice to reserve the PU whose processing rate is closest to the server average is heuristic; a natural test is whether reserving the fastest or slowest PU changes the task completion rate.
  • If the weights $\alpha$ and $\beta$ must be retuned per scenario, the practical value depends on finding a calibration rule; the paper's own suggestion of learning these weights could be tested directly.
  • The suitability-score form is a linear combination of completion time, time margin, and normalized load; other weighting schemes or normalizations could be compared without changing the architecture.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper addresses resource selection in heterogeneous mobile edge computing for autonomous-driving real-time tasks. It proposes SARS, a suitability-based resource selection algorithm that scores each processing unit (PU) using completion time, time margin, and a load factor, and PORA, a proactive reservation mechanism that keeps certain PUs on standby for critical tasks. The experimental section compares SARS and SARS+PORA against classical RSP baselines (shortest-execution, random, latest-PU) across six TSP algorithms and reports task completion rate (TCR) improvements, including up to 13% over the shortest-execution PU baseline when PORA is enabled.

Significance. If the reported improvements are robust, the paper offers a simple, practical addition to the RSP component of task offloading in MEC, which is often under-studied relative to TSP. The algorithms are clearly described with pseudocode, and the paper considers realistic factors such as heterogeneity, distance-dependent transmission delays, and load balancing. However, the quantitative claims as stated are not supported by the current evaluation: the suitability score in Eq. (8) appears to reward high load rather than penalize it, the actual tuned parameters are not reported, the PORA top-k parameter is undefined, and the results lack statistical validation. The central algorithmic idea is defensible, but the evaluation must be corrected and made reproducible before the numeric claims can be accepted.

major comments (4)
  1. [Section 4.1, Eq. (8)] The suitability score in Eq. (8) adds β·lf, where lf is defined in Eq. (7) as the current load divided by the maximum load. Since Algorithm 1 (Line 12) selects the PU with the maximum suitability score, a positive β makes more heavily loaded PUs more attractive, which directly contradicts the text in Section 4.1 stating that the load factor 'ensures that highly utilized PUs are less likely to be allocated to additional tasks, promoting balanced resource usage.' Either Eq. (8) should use −β·lf, or the text must be revised to describe the actual behavior; as written, the algorithm is not reproducible and the claimed load-balancing benefit has no supporting mechanism.
  2. [Section 5.1 and Section 5.2] The paper states in Section 5.1 that α was tuned by trial-and-error within the range 0.5 to 1.5, but it never reports the actual value used in the experiments of Section 5.2. This matters because Eq. (8) can be rewritten as ss = α·td + (1−α)·t_ct + β·lf; the preference between earlier and later completion times reverses at α = 1, so the direction of the selection behavior is unknown without the exact value. Please report the chosen α and β and include a sensitivity analysis to demonstrate that the claimed 13% and 3.88% TCR improvements are not artifacts of a single hand-picked configuration.
  3. [Algorithm 2, Line 11] The parameter k in 'Select the top k PUs with the smallest absolute difference' is never defined in Algorithm 2 or in the accompanying text. Moreover, Line 12 randomly chooses one PU from the top k, introducing nondeterminism without any repeated-run statistics. This makes the PORA component and its contribution to the reported TCR irreproducible; please define k, justify its value, and report results averaged over multiple random choices.
  4. [Section 5.2] All TCR comparisons appear to be based on a single simulation run. Since tasks are randomly generated (Section 5.1) and PORA includes random selection (Algorithm 2, Line 12), the reported differences (1.37%, 1.5%, 3.88%, and 13%) lack error bars, confidence intervals, or any statistical significance test. Please provide mean and variance over multiple independent random seeds and state whether the observed differences are significant.
minor comments (4)
  1. [Section 3.2, Eq. (1)] The notation around Eq. (1) is inconsistent: the text says bwej,k and bweB,u are bandwidths, but later refers to 'bweB,u is the distance between B and ecu.' Please clarify which symbols denote distances and which denote bandwidths, and verify the units of Eq. (1).
  2. [Section 5.1, Table 1] The deadline ranges in Table 1 (e.g., [1,10] for group 1) are difficult to interpret alongside the text that says deadlines are calculated as the release time plus workload plus a random number; please clarify what the table entries represent.
  3. [Figures 2 and 3] The captions of Figs. 2 and 3 do not specify the TSP algorithm used for the comparison or the parameter values of α and β; please add this information to make the figures interpretable.
  4. [Reference [16]] The title of reference [16] contains 'A VE' which appears to be a typographical error for 'AVE' or 'A-VE'; please correct it.

Circularity Check

1 steps flagged · score 6.0 of 10

Headline 13% TCR gain is obtained after trial-and-error tuning of alpha and beta on the same simulation, so the central empirical claim reduces to the fitted scoring weights.

  1. fitted input called prediction [Section 5.1 (Simulation Setup), validating Section 5.2; Eq. (8)]
    "Moreover, we tuned the parameter α in the range of 0.5 to 1.5 and set β to 0.5. Through the trial-and-error method, we found that these parameters effectively weight the calculation of the suitability score in the SARS algorithm."

    The two headline results ("3.88% higher than the latest PU algorithm and 13% higher than the shortest execution PU algorithm", Sec. 5.2) are read off the same simulation on which α and β were tuned by trial and error. Eq. (8), ss = t_ct + α·tm + β·lf with tm = td − t_ct, is algebraically ss = α·td + (1−α)·t_ct + β·lf, so α and β completely determine which PU is selected: α < 1 prefers later-completing PUs and α > 1 prefers earlier-completing PUs. No actual α value is reported, and β = 0.5 makes loaded PUs score higher, contrary to the stated balancing intent. Thus the reported TCR advantage is a property of the tuned weights on the test scenario, not a prediction of a parameter-free algorithm; the empirical claim reduces to the fitted configuration.

full rationale

The algorithmic derivation itself is not circular: Eq. (8) is a heuristic weighted sum, not a rearrangement of the objective max Npt, and SARS is not defined in terms of the TCR metric. PORA's reservation rule is an independent heuristic based on deviation from the average processing rate. The only self-citation (Ref. [13], same authors) is background on an 'emphasized processor' idea and is not load-bearing for the correctness or evaluation of SARS or PORA. There is no imported uniqueness theorem and no ansatz smuggled in through a citation. The one genuine circularity is in the evaluation: α and β are tuned by trial-and-error on the same simulated scenario that produces the up-to-13% comparison, and the actual α value is not reported. This makes the headline empirical result a fitted outcome rather than a parameter-free validation. Separately, Eq. (8)'s positive β·lf term contradicts the text's claim that the load factor makes highly utilized PUs less likely to be chosen; that is an internal-correctness and reproducibility problem, not a circularity, so it does not change the score beyond the fitted-parameter issue.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on two tuned weighting parameters (alpha, beta), an unspecified PORA parameter k, and several domain assumptions about the system model. No new physical entities are introduced.

free parameters (3)
  • alpha = tuned in [0.5, 1.5], final value not reported
    Weight of the time margin in the suitability score (Eq. 8). Tuned by trial-and-error on the same simulation used for the TCR comparison (Section 5.1).
  • beta = 0.5
    Weight of the load factor in the suitability score (Eq. 8). Set by hand; no sensitivity analysis provided (Section 5.1).
  • PORA top-k = unspecified
    In Algorithm 2, Line 11, the algorithm selects the 'top k PUs' with smallest difference from the average rate. The value of k is never defined in the paper, which affects which PU is reserved.
assumptions (3)
  • domain assumption Tasks are non-preemptive and cannot be interrupted once processing starts
    Stated in Section 3.2: a task may be processed on any PU without preemption. This is a common real-time scheduling assumption and is used in the completion-time model.
  • domain assumption Transmission delay is proportional to file size divided by bandwidth multiplied by distance
    Eq. (1) defines transmission delay as (ts_i / bw_e_j,k) * dis_j,k + (ts_i / bw_e_B,u) * dis_B,u. This mixes units and has no physical basis, but it is the model all algorithms are evaluated under.
  • domain assumption The average processing rate of an ECS is a good predictor of which PU should be reserved
    PORA (Algorithm 2) reserves one PU per ECS by choosing randomly among PUs whose rate is closest to the ECS average. The paper provides no theoretical or empirical justification for why this selection is effective.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SARS: A Resource Selection Algorithm for Autonomous Driving Tasks in Heterogeneous Mobile Edge Computing." pith.science (2026). https://pith.science/paper/EJOHSVMQ

@misc{pith2026241115989,
  author       = {Pith},
  title        = {Pith review of: SARS: A Resource Selection Algorithm for Autonomous Driving Tasks in Heterogeneous Mobile Edge Computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EJOHSVMQ}},
  note         = {Machine review of arXiv:2411.15989}
}
read the original abstract

With the rapid advancement of devices requiring intensive computation, such as Internet of Things (IoT) devices, smart sensors, and wearable technology, the computational demands on individual platforms with limited resources have escalated, necessitating the offloading of the generated tasks by the devices to edge. These tasks are often real-time with strict response time requirements. Among these devices, autonomous vehicles present unique challenges due to their critical need for timely and accurate processing to ensure passenger safety. Selecting suitable servers in a heterogeneous mobile edge computing (MEC) architecture is vital to optimizing real-time task processing rates for such applications. To address this, we present an algorithmic solution to improve the allocation of heterogeneous servers to real-time tasks, aiming to maximize the number of processed tasks. By analyzing task and server characteristics in the MEC architecture, we develop the suitability-based adaptive resource selection (SARS) algorithm, which evaluates server suitability based on factors like time constraints and server capabilities. Additionally, we introduce the proactive on-demand resource allocation (PORA) algorithm, which strategically reserves computational resources to ensure availability for critical real-time tasks. We compare the proposed algorithms with several classical and state-of-the-art algorithms. Computational results demonstrate that our approach outperforms existing algorithms, processes more tasks, and effectively prioritizes urgent tasks, particularly in autonomous driving applications.

Figures

Figures reproduced from arXiv: 2411.15989 by the authors.

Figure 1
Figure 1. An illustration of our system architecture. Our architecture features a two-layer hierarchical network. The device layer consists of multiple vehicles at various locations, each equipped with computing platforms that generate tasks from sensor data (e.g., radar, LiDAR, cameras) to perceive their environ￾ment. The MEC layer includes a set of RSUs that relay tasks to a centralized broker, which handles TSP and RSP by … view at source ↗
Figure 2
Figure 2. The TCR of SARS without PORA compared to other RSP algorithms. The difference in TCR values between SARS and other RSP algorithms is noteworthy. The best among the other algorithms is the one that selects the PU capable of finishing the task latest while still meeting the deadline. Despite this, it still underperforms compared to SARS across all TSP algorithms, showing a 1.37% lower TCR in the CR and a 1.5% lower TC… view at source ↗
Figure 3
Figure 3. The TCR comparison of SARS to other RSP algorithms [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: illustrates the impact of SARS as the RSP algorithm on two priority-based scheduling algorithms: ERA and the priority queuing model. We conducted experiments with these priority-based TSP algorithms in two configurations: first, using their original algorithms for both…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages

  1. [1]

    Big Data and Cognitive Comp uting 8(4), Article 42 (2024)

    Garikapati, D., Shetiya, S.S.: Autonomous Vehicles: Evo lution of Artificial Intelligence and the Current Industry Landscape. Big Data and Cognitive Comp uting 8(4), Article 42 (2024)

    work page 2024

  2. [2]

    Journal of Supercomputing 78(1), 1093–1116 (2022)

    Chen, C., Guo, R., Zhang, W., Yang, J., Yeo, C.K.: Optimal s equential relay-remote selection and computation offloading in mobile edge computing. Journal of Supercomputing 78(1), 1093–1116 (2022)

    work page 2022

  3. [3]

    Nature Communications 15(1), Article 4931 (2024)

    Abdel-Aty, M., Ding, S.: A matched case-control analysis of autonomous vs human-driven vehicle accidents. Nature Communications 15(1), Article 4931 (2024)

    work page 2024

  4. [4]

    Journal of Systems Architec ture 94, 14–23 (2019)

    Dai, H., Zeng, X., Yu, Z., Wang, T.: A scheduling algorithm for autonomous driving tasks on mobile edge computing servers. Journal of Systems Architec ture 94, 14–23 (2019)

    work page 2019

  5. [5]

    Sensors 24(12), Article 3948 (2024)

    Wang, Z., Zhao, W., Hu, P., Zhang, X., Liu, L., Fang, C., Sun , Y.: UA V-Assisted Mobile Edge Computing: Dynamic Trajectory Design and Resource Allocat ion. Sensors 24(12), Article 3948 (2024)

    work page 2024

  6. [6]

    IEEE Journal on Selected Areas in Communications 37(5), 1159–1166 (2019)

    Misra, S., Saha, N.: Detour: Dynamic task offloading in soft ware-defined fog for IoT applica- tions. IEEE Journal on Selected Areas in Communications 37(5), 1159–1166 (2019)

    work page 2019

  7. [7]

    Electronics 8(11), Article 1221 (2019)

    Liu, Q., Chen, Z., Wu, J., Deng, Y., Liu, K., Wang, L.: An effic ient task scheduling strat- egy utilizing mobile edge computing in autonomous driving e nvironment. Electronics 8(11), Article 1221 (2019)

    work page 2019

  8. [8]

    Springer Science & Bus iness Media, Vol

    Stankovic, J.A., Spuri, M., Ramamritham, K., Buttazzo, G .: Deadline scheduling for real-time systems: EDF and related algorithms. Springer Science & Bus iness Media, Vol. 460 (1998)

    work page 1998

Show all 30 references
  1. [9]

    Journal of Parallel and D istributed Computing 125, 93– 105 (2019)

    Li, C., Bai, J., Tang, J.: Joint optimization of data place ment and scheduling for improving user experience in edge computing. Journal of Parallel and D istributed Computing 125, 93– 105 (2019)

    2019

  2. [10]

    Inter national Journal of Internet Manu- facturing and Services 7(4), 375–392 (2020)

    Xu, J., Sun, X., Zhang, R., Liang, H., Duan, Q.: Fog-cloud task scheduling of energy con- sumption optimization with deadline consideration. Inter national Journal of Internet Manu- facturing and Services 7(4), 375–392 (2020)

    2020

  3. [11]

    Journal of Network and Computer Applications 204, Article 103333 (2022)

    Azizi, S., Shojafar, M., Abawajy, J., Buyya, R.: Deadlin e-aware and energy-efficient IoT task scheduling in fog computing systems: A semi-greedy app roach. Journal of Network and Computer Applications 204, Article 103333 (2022)

    2022

  4. [12]

    Concur rency and Computation: Prac- tice and Experience 34(22), Article e7148 (2022)

    Gholami, H., Rezvan, M.T.: A cooperative multi-agent offl ine learning algorithm to schedul- ing IoT workflows in the cloud computing environment. Concur rency and Computation: Prac- tice and Experience 34(22), Article e7148 (2022)

    2022

  5. [13]

    In: 2020 6th International Conference on Web Research (ICWR), pp

    Gholami, H., Zakerian, R.: A list-based heuristic algor ithm for static task scheduling in heterogeneous distributed computing systems. In: 2020 6th International Conference on Web Research (ICWR), pp. 21–26. IEEE (2020) SARS: A Resource Selection Algorithm for... 15

    2020

  6. [14]

    Energies 14(6), Article 1788 (2021)

    Balasekaran, G., Jayakumar, S., Pérez de Prado, R.: An in telligent task scheduling mecha- nism for autonomous vehicles via deep learning. Energies 14(6), Article 1788 (2021)

    2021

  7. [15]

    Electronics 12(7), Article 1655 (2023)

    Nie, X., Yan, Y., Zhou, T., Chen, X., Zhang, D.: A Delay-Op timal Task Scheduling Strat- egy for Vehicle Edge Computing Based on the Multi-Agent Deep Reinforcement Learning Approach. Electronics 12(7), Article 1655 (2023)

    2023

  8. [16]

    IEEE Transactions on Vehicular Technology 66(12), 10660– 10675 (2017)

    Feng, J., Liu, Z., Wu, C., Ji, Y.: A VE: Autonomous vehicul ar edge computing framework with ACO-based scheduling. IEEE Transactions on Vehicular Technology 66(12), 10660– 10675 (2017)

    2017

  9. [17]

    In: Pro- ceedings of the ACMSE 2018 Conference, pp

    Choudhari, T., Moh, M., Moh, T.-S.: Prioritized task sch eduling in fog computing. In: Pro- ceedings of the ACMSE 2018 Conference, pp. 1–8. ACM (2018)

    2018

  10. [18]

    Computer Communi cations 172, 75–83 (2021)

    Saba, U.K., Islam, S.u., Ijaz, H., Rodrigues, J.J.P.C., Gani, A., Munir, K.: Planning Fog networks for time-critical IoT requests. Computer Communi cations 172, 75–83 (2021)

    2021

  11. [19]

    In: Proc eedings of the 2024 11th Inter- national Conference on Wireless Communication and Sensor N etworks, pp

    Yu, T.: Optimizing Computational Efficiency in Autonomou s Vehicles: Integrative Edge and Cloud Computing Strategies in Vehicular Networks. In: Proc eedings of the 2024 11th Inter- national Conference on Wireless Communication and Sensor N etworks, pp. 5–12 (2024)

    2024

  12. [20]

    IEEE Internet of Th ings Journal 7(6), 4961–4971 (2020)

    Liu, Y., Wang, S., Zhao, Q., Du, S., Zhou, A., Ma, X., Yang, F.: Dependency-aware task scheduling in vehicular edge computing. IEEE Internet of Th ings Journal 7(6), 4961–4971 (2020)

    2020

  13. [21]

    IEEE Transactions on Computers 53, 1462–1473 (2004)

    Bini, E., Buttazzo, G.C.: Schedulability analysis of pe riodic fixed priority systems. IEEE Transactions on Computers 53, 1462–1473 (2004)

    2004

  14. [22]

    The Journal of Superc omputing, 1–33 (2024)

    Lin, J., Rao, H., Liang, S., Zhao, Y., Ren, Q., Jia, G.: Aph to: a task offloading strategy for autonomous driving under mobile edge. The Journal of Superc omputing, 1–33 (2024)

    2024

  15. [23]

    Peer-to-Peer Netwo rking and Applications 11, 793– 807 (2018)

    Wang, T., Wei, X., Tang, C., Fan, J.: Efficient multi-tasks scheduling algorithm in mobile cloud computing with time constraints. Peer-to-Peer Netwo rking and Applications 11, 793– 807 (2018)

    2018

  16. [24]

    The Jour nal of Supercomputing 77(11), 12486–12507 (2021)

    Li, W., Jin, S.: Performance evaluation and optimizatio n of a task offloading strategy on the mobile edge computing with edge heterogeneity. The Jour nal of Supercomputing 77(11), 12486–12507 (2021)

    2021

  17. [25]

    The Journal of Superc omputing 79(9), 10139–10177 (2023)

    Zhao, H., Geng, J., Jin, S.: Performance research on a tas k offloading strategy in a two- tier edge structure-based MEC system. The Journal of Superc omputing 79(9), 10139–10177 (2023)

    2023

  18. [26]

    IEEE Internet of Things Journal 5(1), 283–294 (2018)

    Liu, L., Chang, Z., Guo, X., Mao, S., Ristaniemi, T.: Mult iobjective optimization for com- putation offloading in fog computing. IEEE Internet of Things Journal 5(1), 283–294 (2018)

    2018

  19. [27]

    Future Generation Computer Systems 133, 228–239 (2022)

    Liu, J., Zhang, X.: Truthful resource trading for depend ent task offloading in heterogeneous edge computing. Future Generation Computer Systems 133, 228–239 (2022)

    2022

  20. [28]

    arXiv preprint arXiv:2403.15665 (2024)

    Rublein, C., Mehmeti, F., Mahon, M., La Porta, T.F.: Impr oved methods of task assign- ment and resource allocation with preemption in edge comput ing systems. arXiv preprint arXiv:2403.15665 (2024)

    2024 arXiv

  21. [29]

    F uture Generation Computer Sys- tems 150, 243–254 (2024)

    Jeremiah, S.R., Yang, L.T., Park, J.H.: Digital twin-as sisted resource allocation framework based on edge collaboration for vehicular edge computing. F uture Generation Computer Sys- tems 150, 243–254 (2024)

    2024

  22. [30]

    Future Generation Computer Systems 128, 132–141 (2022)

    Guo, K., Zhang, R.: Fairness-oriented computation offloa ding for cloud-assisted edge com- puting. Future Generation Computer Systems 128, 132–141 (2022)

    2022

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.