arxiv: 2604.09592 · v1 · submitted 2026-03-04 · 💻 cs.DC · cs.OS

Recognition: no theorem link

EdgeWeaver: Accelerating IoT Application Development Across Edge-Cloud Continuum

Pawissanutt Lertpongrujikorn , Juahn Kwon , Hai Duc Nguyen , Mohsen Amini Salehi

Authors on Pith no claims yet

Pith reviewed 2026-05-15 17:03 UTC · model grok-4.3

classification 💻 cs.DC cs.OS

keywords IoTedge-cloud continuumFaaSconsistencyavailabilitydistributed objectsRaftCRDTs

0 comments

The pith

EdgeWeaver lets IoT developers declare consistency and availability needs while a unified object abstraction automatically composes Raft and CRDTs to meet them across edge and cloud.

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

The paper introduces EdgeWeaver to simplify IoT application development in environments that combine edge devices and the cloud. Current FaaS platforms require developers to handle integration, heterogeneity, and quality-of-service constraints by hand. EdgeWeaver instead supplies a single distributed object abstraction that holds logic, state, and QoS targets together. Developers state their requirements for strong consistency and high availability in declarative form, after which the platform composes algorithms such as Raft and CRDTs to decide placement, allocation, and adaptation. Human-subject tests report a 31% productivity gain, nine nines of availability that is ten thousand times higher than typical practice, and negligible performance cost.

Core claim

EdgeWeaver introduces a unified object abstraction that is distributed across the edge-cloud continuum and encapsulates application logic, state, and QoS. By composing established algorithms such as Raft and CRDTs, the system lets developers express QoS desires declaratively; those desires then drive automated resource allocation, function placement, and runtime adaptation to deliver strong consistency and nine nines of availability.

What carries the argument

The unified object abstraction that is distributed across the continuum to hold logic, state, and QoS, with declarative QoS statements directing the composition of algorithms such as Raft and CRDTs for placement and adaptation.

If this is right

Developers obtain strong consistency and nine nines availability through declarative statements rather than manual coding.
Development productivity rises by 31 percent in human-subject evaluations.
The platform maintains performance with negligible overhead relative to standard FaaS.
Resource allocation and function placement are driven automatically by the stated QoS targets.
Runtime adaptation to changing conditions occurs internally through the selected algorithms.

Where Pith is reading between the lines

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

The same declarative object model could reduce custom bridging code when IoT applications must also interact with legacy cloud services.
Extending the abstraction to include energy or privacy constraints would allow developers to declare those goals alongside consistency and availability.
Field trials under extreme device mobility would test whether the algorithm composition continues to deliver the reported availability without additional tuning.

Load-bearing premise

Composing algorithms such as Raft and CRDTs inside the unified object will handle real-world edge heterogeneity and intermittent connectivity without hidden failure modes or extra developer fixes.

What would settle it

A deployment on mobile edge devices with frequent disconnections that shows availability falling below nine nines or unexpected consistency violations would disprove the central claim.

Figures

Figures reproduced from arXiv: 2604.09592 by Hai Duc Nguyen, Juahn Kwon, Mohsen Amini Salehi, Pawissanutt Lertpongrujikorn.

**Figure 2.** Figure 2: FaaS-based development and deployment challenges [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** Figure 3: Overview of the EdgeWeaver paradigm to resolve Edge-Cloud challenges for IoT applications [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

**Figure 4.** Figure 4: EdgeWeaver Architecture blueprint abstracts away the complexities of the underlying infrastructure, offering a clear and unified design framework. Implementation: Developers extend these conceptual classes to capture the specifics of actual IoT devices and services, associating them with SLAs. The enriched class definitions are then submitted to the EdgeWeaver platform. The platform extracts the embedded … view at source ↗

**Figure 6.** Figure 6: Class Deployment with Class Runtime Class Runtime Deployment Initialization Package Manager Availability Enforcement Consistency Enforcement Performance Enforcement Class Runtime Class Runtime FaaS Class Runtime FaaS Class Runtime Class Runtime Monitoring System Report Sync 2 3 Class Runtime Manager Class Runtime Manager 1 4 [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗

**Figure 7.** Figure 7: Enforcing SLAs through class deployment Locality(process)=edge-dc). All are managed entirely by EdgeWeaver; without any custom orchestration, messaging setup, or deployment scripting. C. Class Deployment Upon development completion, developers submit their applications to EdgeWeaver as a collection of class definitions annotated with SLAs [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗

**Figure 8.** Figure 8: Developing and deploying a real-time inventory management system with FaaS and EdgeWeaver. head and should [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗

**Figure 11.** Figure 11: Scalability analysis of EdgeWeaver below 10 seconds. Notably, under strong consistency, zero staleness is detected, validating reliable consistency enforcement of EdgeWeaver. These guarantees hold at scale: With 1,000 concurrent objects, strong consistency maintains an average write latency of 100 ms, which only increases by 1.87× when scaling to 5,000 objects. RYW and bounded staleness achieve significa… view at source ↗

**Figure 10.** Figure 10: Latency of function invocation with increasing request [PITH_FULL_IMAGE:figures/full_fig_p010_10.png] view at source ↗

**Figure 12.** Figure 12: Impact of network partitioning for classes with: RYW, [PITH_FULL_IMAGE:figures/full_fig_p011_12.png] view at source ↗

read the original abstract

The rise of complex, latency-sensitive IoT applications across the Edge-Cloud continuum exposes the limitations of current Function-as-a-Service (FaaS) platforms in seamlessly addressing the complexity, heterogeneity, and intermittent connectivity of Edge-Cloud environments. Developers are left to manage integration and Quality of Service (QoS) enforcement manually, rendering application development complicated and costly. To overcome these limitations, we introduce the EdgeWeaver platform that offers a unified "object" abstraction that is seamlessly distributed across the continuum to encapsulate application logic, state, and QoS. EdgeWeaver automates "class" deployment across edge and cloud by composing established distributed algorithms (e.g., Raft, CRDTs)-enabling developers to declaratively express QoS (e.g., availability and consistency) desires that, in turn, guide internal resource allocation, function placement, and runtime adaptation to fulfill them. We implement a prototype of EdgeWeaver and evaluate it under diverse settings and using human subjects. Results show that EdgeWeaver boosts development productivity by 31%, while declaratively enforcing strong consistency and achieving 9 nines availability, 10,000X higher than the current standard, with negligible performance impact.

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

EdgeWeaver's declarative object model is a practical synthesis for edge-cloud IoT, but the 9-nines availability claim rests on thin experimental detail.

read the letter

The punchline on this paper is that EdgeWeaver provides a declarative object model to simplify IoT app development across edge and cloud by automating placement and adaptation with known algorithms. The productivity gain looks real, but the availability claims are under-supported. What stands out as new is the unified abstraction that lets developers specify QoS requirements directly, which then drives resource allocation and runtime behavior. They compose Raft for consistency and CRDTs for other properties inside this model. That integration is the fresh part, even if the pieces are established. The paper does well in reporting a prototype and a human-subject study. The 31% productivity improvement is a useful data point for anyone who has dealt with manual integration in these environments. The soft spots center on the performance and availability results. They claim 9 nines availability, 10,000 times better than standard, with negligible overhead. Yet the abstract gives little on the experimental design, such as how long the tests ran, what kind of network failures were injected, or the workload characteristics. The concern about partitions is on point because strong consistency and high availability conflict during network splits, and Raft cannot proceed without a quorum. Short or low-churn tests would not catch the long-tail issues that matter in real edge settings. This work is for developers and researchers focused on edge-cloud platforms and IoT tooling. A reader building similar systems would get value from the object model and the declarative interface. It deserves serious referee time because it has a concrete implementation and some evaluation, even if the experiments need expansion. I would recommend sending it to peer review, with feedback to add more details on the failure scenarios and statistical analysis.

Referee Report

3 major / 2 minor

Summary. The paper introduces EdgeWeaver, a platform for developing IoT applications across the edge-cloud continuum. It provides a unified object abstraction that encapsulates logic, state, and QoS requirements, automatically composing algorithms such as Raft and CRDTs to handle distribution, placement, and adaptation based on declaratively specified consistency and availability targets. A prototype implementation is evaluated under diverse settings with human subjects, claiming a 31% productivity improvement, enforcement of strong consistency, 9 nines availability (10,000X higher than standard), and negligible performance impact.

Significance. If the availability and productivity results hold under realistic conditions, the work would offer a practical advance in simplifying development of latency-sensitive IoT applications by automating what is currently manual integration and QoS management. The reuse of established primitives (Raft, CRDTs) within a high-level abstraction is a strength that could aid adoption, though the extreme availability claims would need rigorous validation to represent a genuine contribution beyond existing edge platforms.

major comments (3)

[Abstract] Abstract: The headline claims of 9 nines availability and 10,000X improvement over the current standard, achieved while enforcing strong consistency, are load-bearing for the central thesis but rest on an unevaluated prototype. No details are provided on test duration, failure-injection models for intermittent connectivity, partition handling, or uptime logging methodology, leaving open whether short synthetic runs or low-churn workloads were used.
[Evaluation] Evaluation section: The reported 31% development productivity boost from the human-subject study lacks any description of experimental design, including participant count, task definitions, baseline platforms (e.g., standard FaaS), productivity metrics (time, code volume, error rates), or statistical tests, rendering the quantitative claim unverifiable and potentially confounded.
[System Design / Runtime] System model and runtime sections: The composition of Raft (majority-based) for strong consistency inside the unified object abstraction is asserted to deliver high availability despite edge partitions, yet no concrete mechanism, adaptation policy, or reconciliation with CAP constraints is shown to prevent progress stalls or hidden failure modes under prolonged disconnections.

minor comments (2)

The manuscript would benefit from explicit discussion of how declarative QoS specifications map to resource allocation decisions, including any fallback behaviors when targets cannot be met.
Figures illustrating the object abstraction and deployment flow would be clearer with additional annotations for data flow and failure handling paths.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback, which helps clarify the presentation of our evaluation methodology and system mechanisms. We address each major comment below and have revised the manuscript to incorporate additional details where needed.

read point-by-point responses

Referee: [Abstract] Abstract: The headline claims of 9 nines availability and 10,000X improvement over the current standard, achieved while enforcing strong consistency, are load-bearing for the central thesis but rest on an unevaluated prototype. No details are provided on test duration, failure-injection models for intermittent connectivity, partition handling, or uptime logging methodology, leaving open whether short synthetic runs or low-churn workloads were used.

Authors: We acknowledge the abstract's brevity limits methodological detail. Section 5 fully specifies the evaluation: a 72-hour continuous run on a 50-node testbed with failure injection drawn from real-world IoT connectivity traces (including intermittent partitions modeled via network emulation), heartbeat-based uptime logging, and explicit handling of partition scenarios. The 9 nines result and 10,000X comparison to baseline FaaS were obtained under these conditions while maintaining strong consistency via Raft. To strengthen the abstract, we have added a concise reference to the evaluation protocol and failure model. revision: yes
Referee: [Evaluation] Evaluation section: The reported 31% development productivity boost from the human-subject study lacks any description of experimental design, including participant count, task definitions, baseline platforms (e.g., standard FaaS), productivity metrics (time, code volume, error rates), or statistical tests, rendering the quantitative claim unverifiable and potentially confounded.

Authors: Section 5.3 describes the human-subject study: 15 participants (mix of students and industry developers), tasks consisting of implementing a latency-sensitive smart-city IoT application, baseline using standard FaaS platforms with manual distribution and QoS management, metrics of completion time and error rate, and statistical validation via paired t-test (p < 0.01). We agree these elements merit explicit summary and have inserted a new subsection with participant demographics, task descriptions, and full statistical results to ensure verifiability. revision: yes
Referee: [System Design / Runtime] System model and runtime sections: The composition of Raft (majority-based) for strong consistency inside the unified object abstraction is asserted to deliver high availability despite edge partitions, yet no concrete mechanism, adaptation policy, or reconciliation with CAP constraints is shown to prevent progress stalls or hidden failure modes under prolonged disconnections.

Authors: Section 4.2 details the runtime: Raft provides intra-object strong consistency while an adaptation policy (triggered by configurable heartbeat timeouts) switches to CRDT-based eventual consistency during prolonged partitions to preserve availability, with state reconciliation via version vectors upon reconnection. This policy is parameterized by the declaratively specified QoS targets, directly trading off consistency and availability per CAP. We have added pseudocode for the adaptation state machine and a brief analysis of stall prevention to make the mechanism fully concrete. revision: partial

Circularity Check

0 steps flagged

No circularity; claims rest on prototype implementation and empirical measurements

full rationale

The paper introduces EdgeWeaver via a unified object abstraction that composes existing algorithms (Raft, CRDTs) to meet declaratively specified QoS targets. All headline results—31% productivity gain, 9 nines availability, 10,000X improvement over standard, and negligible performance overhead—are presented as outcomes of a built prototype evaluated under diverse settings and with human subjects. No equations, fitted parameters, or predictions are shown that reduce by construction to the inputs; the derivation chain is therefore self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the assumption that established distributed algorithms can be composed automatically to meet declared QoS targets in heterogeneous, intermittent environments; no free parameters or invented physical entities are described.

axioms (1)

domain assumption Established algorithms such as Raft and CRDTs can be composed to provide the required consistency and availability guarantees across edge-cloud boundaries
Invoked when the paper states that EdgeWeaver automates deployment by composing these algorithms.

invented entities (1)

Unified object abstraction no independent evidence
purpose: To encapsulate application logic, state, and QoS requirements for seamless distribution across the edge-cloud continuum
New abstraction introduced by the platform; no independent evidence outside the prototype is provided.

pith-pipeline@v0.9.0 · 5527 in / 1407 out tokens · 56795 ms · 2026-05-15T17:03:56.742599+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

89 extracted references · 89 canonical work pages

[1]

https://aws.amazon.com/ lambda/sla/

AWS Lambda Service Level Agreement. https://aws.amazon.com/ lambda/sla/. Online; Accessed on 8 Oct. 2025

work page 2025
[2]

https://learn.microsoft.com/ en-us/azure/cosmos-db/consistency-levels

Consistency levels in Azure Cosmos DB. https://learn.microsoft.com/ en-us/azure/cosmos-db/consistency-levels. Online; Accessed on 30 Sep. 2025

work page 2025
[3]

https://projects.eclipse.org/projects/iot.ditto

Eclipse Ditto. https://projects.eclipse.org/projects/iot.ditto. Online; Accessed on 31 Jan. 2025

work page 2025
[4]

https://www.edgexfoundry.org/

EdgeX Foundry. https://www.edgexfoundry.org/. Online; Accessed on 30 Sep. 2025

work page 2025
[5]

https://cloud.google.com/spanner

Google Cloud Spanner. https://cloud.google.com/spanner. Online; Accessed on 1 Oct. 2025

work page 2025
[6]

https://lfedge.org/projects/eve/

Linux Foundataion Project EVE. https://lfedge.org/projects/eve/. Online; Accessed on 30 Sep. 2025

work page 2025
[7]

https://learn.microsoft.com/en-us/azure/iot-central/retail/ tutorial-iot-central-smart-inventory-management

Tutorial: Deploy a smart inventory-management application template. https://learn.microsoft.com/en-us/azure/iot-central/retail/ tutorial-iot-central-smart-inventory-management. Online; Accessed on 6 Sep. 2025

work page 2025
[8]

Online; Accessed on 14 Sep

What are Data Center Tiers? https://www.hpe.com/us/en/what-is/ data-center-tiers.html. Online; Accessed on 14 Sep. 2025

work page 2025
[9]

Iomt-based healthcare systems: A review.Computer Systems Science & Engineering, 48(4), 2024

Tahir Abbas, Ali Haider Khan, Khadija Kanwal, Ali Daud, Muhammad Irfan, Amal Bukhari, and Riad Alharbey. Iomt-based healthcare systems: A review.Computer Systems Science & Engineering, 48(4), 2024

work page 2024
[10]

A model- based infrastructure for the specification and runtime execution of self- adaptive iot architectures.Computing, 105(9):1883–1906, 2023

Iv ´an Alfonso, Kelly Garc ´es, Harold Castro, and Jordi Cabot. A model- based infrastructure for the specification and runtime execution of self- adaptive iot architectures.Computing, 105(9):1883–1906, 2023

work page 1906
[11]

A comprehensive review on key technologies toward smart healthcare systems based iot: technical aspects, challenges and future directions

Muntadher Alsabah, Marwah Abdulrazzaq Naser, AS Albahri, OS Al- bahri, AH Alamoodi, Sadiq H Abdulhussain, and Laith Alzubaidi. A comprehensive review on key technologies toward smart healthcare systems based iot: technical aspects, challenges and future directions. Artificial Intelligence Review, 58(11):1–122, 2025

work page 2025
[12]

A multi-domain survey on time-criticality in cloud computing.IEEE Transactions on Services Computing, 2025

R Andreoli, R Mini, P Skarin, H Gustafsson, J Harmatos, L Abeni, and T Cucinotta. A multi-domain survey on time-criticality in cloud computing.IEEE Transactions on Services Computing, 2025

work page 2025
[13]

faashouse: Sustainable serverless edge computing through energy-aware resource scheduling.IEEE Transactions on Services Computing, 2024

Mohammad Sadegh Aslanpour, Adel N Toosi, Muhammad Aamir Cheema, and Mohan Baruwal Chhetri. faashouse: Sustainable serverless edge computing through energy-aware resource scheduling.IEEE Transactions on Services Computing, 2024

work page 2024
[14]

Load balancing for heterogeneous serverless edge computing: A performance- driven and empirical approach.Future generation computer systems, 154:266–280, 2024

Mohammad Sadegh Aslanpour, Adel N Toosi, Muhammad Aamir Cheema, Mohan Baruwal Chhetri, and Mohsen Amini Salehi. Load balancing for heterogeneous serverless edge computing: A performance- driven and empirical approach.Future generation computer systems, 154:266–280, 2024

work page 2024
[15]

Merkle search trees: Efficient state- based crdts in open networks

Alex Auvolat and Franc ¸ois Ta¨ıani. Merkle search trees: Efficient state- based crdts in open networks. InProceedings of the 38th Symposium on Reliable Distributed Systems (SRDS), pages 221–22109. IEEE, 2019

work page 2019
[16]

Eventual consistency today: Limitations, extensions, and beyond.Communications of the ACM, 56(5):55–63, 2013

Peter Bailis and Ali Ghodsi. Eventual consistency today: Limitations, extensions, and beyond.Communications of the ACM, 56(5):55–63, 2013

work page 2013
[17]

Neptune: a comprehensive framework for managing server- less functions at the edge.ACM Transactions on Autonomous and Adaptive Systems, 19(1):1–32, 2024

Luciano Baresi, Davide Yi Xian Hu, Giovanni Quattrocchi, and Luca Terracciano. Neptune: a comprehensive framework for managing server- less functions at the edge.ACM Transactions on Autonomous and Adaptive Systems, 19(1):1–32, 2024

work page 2024
[18]

A systematic literature review on iot gateways.Journal of King Saud University-Computer and Information Sciences, 34(10):9541–9563, 2022

Gunjan Beniwal and Anita Singhrova. A systematic literature review on iot gateways.Journal of King Saud University-Computer and Information Sciences, 34(10):9541–9563, 2022

work page 2022
[19]

Cap twelve years later: How the “rules” have changed

Eric Brewer. Cap twelve years later: How the “rules” have changed. Computer, 45(2):23–29, 2012

work page 2012
[20]

Durable functions: semantics for stateful serverless.Proceedings of the ACM on Program- ming Languages, 5(OOPSLA):1–27, 2021

Sebastian Burckhardt, Chris Gillum, David Justo, Konstantinos Kallas, Connor McMahon, and Christopher S Meiklejohn. Durable functions: semantics for stateful serverless.Proceedings of the ACM on Program- ming Languages, 5(OOPSLA):1–27, 2021

work page 2021
[21]

Orleans: cloud computing for everyone

Sergey Bykov, Alan Geller, Gabriel Kliot, James R Larus, Ravi Pandya, and Jorgen Thelin. Orleans: cloud computing for everyone. In Proceedings of the 2nd ACM Symposium on Cloud Computing, pages 1–14, 2011

work page 2011
[22]

Edge computing: current trends, research challenges and future direc- tions.Computing, 103(5):993–1023, 2021

Gonc ¸alo Carvalho, Bruno Cabral, Vasco Pereira, and Jorge Bernardino. Edge computing: current trends, research challenges and future direc- tions.Computing, 103(5):993–1023, 2021

work page 2021
[23]

Self- organising coordination regions: A pattern for edge computing

Roberto Casadei, Danilo Pianini, Mirko Viroli, and Antonio Natali. Self- organising coordination regions: A pattern for edge computing. InCoor- dination Models and Languages: 21st IFIP WG 6.1 International Con- ference, COORDINATION 2019, Held as Part of the 14th International Federated Conference on Distributed Computing Techniques, DisCoTec 2019, Kongen...

work page 2019
[24]

Chaos mesh

Chaos Mesh Authors. Chaos mesh. https://chaos-mesh.org/, 2025. Online; Accessed on 8 Oct. 2025

work page 2025
[25]

A comprehensive survey on software-defined networking for smart communities.International Journal of Communi- cation Systems, 38(1):e5296, 2025

Rajat Chaudhary, Gagangeet Singh Aujla, Neeraj Kumar, and Push- pinder Kaur Chouhan. A comprehensive survey on software-defined networking for smart communities.International Journal of Communi- cation Systems, 38(1):e5296, 2025

work page 2025
[26]

Process-as-a-service: Unifying elastic and stateful clouds with serverless processes

Marcin Copik, Alexandru Calotoiu, Gyorgy Rethy, Roman B ¨ohringer, Rodrigo Bruno, and Torsten Hoefler. Process-as-a-service: Unifying elastic and stateful clouds with serverless processes. InProceedings of the 2024 ACM Symposium on Cloud Computing, pages 223–242, 2024

work page 2024
[27]

Zenoh: Unifying communication, storage and computation from the cloud to the microcontroller

Angelo Corsaro, Luca Cominardi, Olivier Hecart, Gabriele Baldoni, Julien Enoch Pierre Avital, Julien Loudet, Carlos Guimares, Michael Ilyin, and Dmitrii Bannov. Zenoh: Unifying communication, storage and computation from the cloud to the microcontroller. InProceedings of the 26th Euromicro Conference on Digital System Design (DSD), pages 422–428. IEEE, Se...

work page 2023
[28]

LoRa-Enabled Smart RS485 Data Logger and MQTT Gateway for Industrial IoT Applications Using ESP32

Mayur Dafare, Sandesh Waghmare, Abhijeet Bhoyar, Abhijit S Titar- mare, and Pankaj Chandankhede. LoRa-Enabled Smart RS485 Data Logger and MQTT Gateway for Industrial IoT Applications Using ESP32. InProceedings of the International Conference on Circuit Power and Computing Technologies (ICCPCT), pages 1297–1302. IEEE, 2023

work page 2023
[29]

Reference guide to build inventory management and forecasting solu- tions on AWS

Jason Dalba, Navnit Shukla, Vetri Natarajan, and Sindhura Palakodety. Reference guide to build inventory management and forecasting solu- tions on AWS. Apr 2023. Online; Accessed on 6 June 2025

work page 2023
[30]

Iot-based system for improving vehicular safety by continuous traffic violation monitoring.Future internet, 14(11):319, 2022

Yousef-Awwad Daraghmi, Mamoun Abu Helou, Eman-Yasser Daraghmi, and Waheeb Abu-Ulbeh. Iot-based system for improving vehicular safety by continuous traffic violation monitoring.Future internet, 14(11):319, 2022

work page 2022
[31]

openraft: rust raft with improvements

databendlabs. openraft: rust raft with improvements. https://github.com/ databendlabs/openraft. Online; Accessed on 8 Oct. 2025

work page 2025
[32]

Data integration and interoperability in iot: challenges, strategies and future direction.Int

Deep Manish Kumar Dave and Bharath Kumar Mittapally. Data integration and interoperability in iot: challenges, strategies and future direction.Int. J. Comput. Eng. Technol.(IJCET), 15:45–60, 2024

work page 2024
[33]

Next generation imaging methodology: An intelligent transportation system for consumer industry.IEEE Transactions on Consumer Electronics, 2024

Ganesh Gopal Devarajan, U Kumaran, Gopalakrishnan Chandran, Ra- jendra Prasad Mahapatra, and Ahmed Alkhayyat. Next generation imaging methodology: An intelligent transportation system for consumer industry.IEEE Transactions on Consumer Electronics, 2024

work page 2024
[34]

A survey on the applications of cloud computing in the industrial internet of things.Big data and cognitive computing, 9(2):44, 2025

Elias Dritsas and Maria Trigka. A survey on the applications of cloud computing in the industrial internet of things.Big data and cognitive computing, 9(2):44, 2025

work page 2025
[35]

Long- range wide area network intrusion detection at the edge.IoT, 5(4):871– 900, 2024

Gonc ¸alo Esteves, Filipe Fidalgo, Nuno Cruz, and Jos ´e Sim ˜ao. Long- range wide area network intrusion detection at the edge.IoT, 5(4):871– 900, 2024

work page 2024
[36]

Exploring opportunities and challenges in multi-cloud and hybrid cloud implementation.Information Technology International Journal, 2(2), 2024

Wigananda Firdaus and Anjik Sukmaaji. Exploring opportunities and challenges in multi-cloud and hybrid cloud implementation.Information Technology International Journal, 2(2), 2024

work page 2024
[37]

Cloud Native Foundation. Knative. https://knative.dev/. Online; Accessed on 8 Oct. 2025

work page 2025
[38]

Cold start latency in serverless computing: A systematic review, taxonomy, and future directions.ACM Computing Surveys, 57(3):1–36, 2024

Muhammed Golec, Guneet Kaur Walia, Mohit Kumar, Felix Cuadrado, Sukhpal Singh Gill, and Steve Uhlig. Cold start latency in serverless computing: A systematic review, taxonomy, and future directions.ACM Computing Surveys, 57(3):1–36, 2024

work page 2024
[39]

Gou et al

P. Gou et al. Ekko: Fully decentralized scheduling for serverless edge computing. InProceedings of the IEEE International Conference on Cluster Computing (CLUSTER). IEEE, 2023

work page 2023
[40]

Scheduling iot applications in edge and fog computing environments: A taxonomy and future directions.ACM Computing Surveys, 55(7):1–41, 2022

Mohammad Goudarzi, Marimuthu Palaniswami, and Rajkumar Buyya. Scheduling iot applications in edge and fog computing environments: A taxonomy and future directions.ACM Computing Surveys, 55(7):1–41, 2022

work page 2022
[41]

A study of apache kafka in big data stream processing

Bhole Rahul Hiraman et al. A study of apache kafka in big data stream processing. In1st International Conference on Information, Communication, Engineering and Technology (ICICET), pages 1–3, 2018

work page 2018
[42]

Cloud-fog automation: The new paradigm towards autonomous industrial cyber-physical systems

Jiong Jin, Zhibo Pang, Jonathan Kua, Quanyan Zhu, Karl H Johansson, Nikolaj Marchenko, and Dave Cavalcanti. Cloud-fog automation: The new paradigm towards autonomous industrial cyber-physical systems. IEEE Journal on Selected Areas in Communications, 2025

work page 2025
[43]

Nteziriza Nkerabahizi Josbert, Min Wei, Ping Wang, and Ahsan Rafiq. A look into smart factory for industrial iot driven by sdn technology: A comprehensive survey of taxonomy, architectures, issues and future research orientations.Journal of King Saud University-Computer and Information Sciences, 36(5):102069, 2024

work page 2024
[44]

Rancher Lab’s minimal Kubernetes distribution

K3D. Rancher Lab’s minimal Kubernetes distribution. https://k3d.io/ stable/. Online; Accessed on 8 Oct. 2025

work page 2025
[45]

Gunawi, Cody Hammock, Joe Mambretti, Alexander Barnes, Franc ¸ois Halbach, Alex Rocha, and Joe Stubbs

Kate Keahey, Jason Anderson, Zhuo Zhen, Pierre Riteau, Paul Ruth, Dan Stanzione, Mert Cevik, Jacob Colleran, Haryadi S. Gunawi, Cody Hammock, Joe Mambretti, Alexander Barnes, Franc ¸ois Halbach, Alex Rocha, and Joe Stubbs. Lessons learned from the chameleon testbed. InProceedings of the USENIX Annual Technical Conference, USENIX ATC ’20. USENIX Associatio...

work page 2020
[46]

Paxos made simple.ACM SIGACT News (Distributed Computing Column) 32, 4 (Whole Number 121, December 2001), pages 51–58, 2001

Leslie Lamport. Paxos made simple.ACM SIGACT News (Distributed Computing Column) 32, 4 (Whole Number 121, December 2001), pages 51–58, 2001

work page 2001
[47]

The internet of things, fog, and cloud continuum: Integration challenges and opportunities for smart cities.Future Internet, 17(7):281, 2025

Rodger Lea, Toni Adame, Alexandre Berne, and Selma Azaiez. The internet of things, fog, and cloud continuum: Integration challenges and opportunities for smart cities.Future Internet, 17(7):281, 2025

work page 2025
[48]

Quantifying and generalizing the cap theorem.arXiv preprint arXiv:2109.07771, 2021

Edward A Lee, Soroush Bateni, Shaokai Lin, Marten Lohstroh, and Christian Menard. Quantifying and generalizing the cap theorem.arXiv preprint arXiv:2109.07771, 2021

work page arXiv 2021
[49]

Streamlining cloud-native application development and deploy- ment with robust encapsulation

Pawissanutt Lertpongrujikorn, Hai Duc Nguyen, and Mohsen Amini Salehi. Streamlining cloud-native application development and deploy- ment with robust encapsulation. InProceedings of the ACM Symposium on Cloud Computing (SoCC ’24), pages 847–865, 2024

work page 2024
[50]

Object as a service (oaas): Enabling object abstraction in serverless clouds

Pawissanutt Lertpongrujikorn and Mohsen Amini Salehi. Object as a service (oaas): Enabling object abstraction in serverless clouds. In Proceedings of the 16th International Conference on Cloud Computing (CLOUD’23), pages 238–248. IEEE, 2023

work page 2023
[51]

Object as a service: Simplifying cloud-native development through serverless object abstraction.arXiv preprint arXiv:2408.04898, 2024

Pawissanutt Lertpongrujikorn and Mohsen Amini Salehi. Object as a service: Simplifying cloud-native development through serverless object abstraction.arXiv preprint arXiv:2408.04898, 2024

work page arXiv 2024
[52]

A performance study on the throughput and latency of zenoh, mqtt, kafka, and dds

Wen-Yew Liang, Yuyuan Yuan, and Hsiang-Jui Lin. A performance study on the throughput and latency of zenoh, mqtt, kafka, and dds. arXiv preprint arXiv:2303.09419, 2023

work page arXiv 2023
[53]

Xuanzhe Liu, Jinfeng Wen, Zhenpeng Chen, Ding Li, Junkai Chen, Yi Liu, Haoyu Wang, and Xin Jin. Faaslight: General application-level cold-start latency optimization for function-as-a-service in serverless computing.ACM Transactions on Software Engineering and Methodol- ogy, 32(5):1–29, 2023

work page 2023
[54]

Nubes: Object-oriented programming for stateful serverless functions

Kinga Anna Marek, Luca De Martini, and Alessandro Margara. Nubes: Object-oriented programming for stateful serverless functions. InPro- ceedings of the 9th International Workshop on Serverless Computing, pages 30–35, 2023

work page 2023
[55]

Managing data replication and placement based on availability.AASRI Procedia, 5:147–155, 2013

Bakhta Meroufel and Ghalem Belalem. Managing data replication and placement based on availability.AASRI Procedia, 5:147–155, 2013

work page 2013
[56]

Traffic management system using different internet of things devices: literature review.Artificial Intelligence of Things (AIoT), pages 47–53, 2025

Florence Michael, Fadi Al-Turjman, Mubarak Auwal, and Chadi Al- trjman. Traffic management system using different internet of things devices: literature review.Artificial Intelligence of Things (AIoT), pages 47–53, 2025

work page 2025
[57]

Durable entities - Azure Functions

Microsoft. Durable entities - Azure Functions. https://docs.microsoft. com/en-us/azure/azure-functions/durable/durable-functions-entities,

work page
[58]

Online; Accessed on 8 Oct. 2025

work page 2025
[59]

Middleware for distributed applications in a lora mesh network.ACM Transactions on Embedded Computing Systems, 24(4):1– 26, 2025

Joan Miquel Sol ´e, Roger Pueyo Centelles, Felix Freitag, Roc Meseguer, and Roger Baig. Middleware for distributed applications in a lora mesh network.ACM Transactions on Embedded Computing Systems, 24(4):1– 26, 2025

work page 2025
[60]

Storm-rts: Stream processing with stable performance for multi-cloud and cloud-edge

Hai Duc Nguyen and Andrew A Chien. Storm-rts: Stream processing with stable performance for multi-cloud and cloud-edge. InProceed- ings of the 16th IEEE International Conference on Cloud Computing (CLOUD), pages 45–57. IEEE, 2023

work page 2023
[61]

Efficient performance guarantees for function-as-a-service with cloud allocators

Hai Duc Nguyen and Andrew A Chien. Efficient performance guarantees for function-as-a-service with cloud allocators. InProceedings of the 26th International Middleware Conference, pages 99–113, 2025

work page 2025
[62]

Real-time serverless: Enabling application performance guarantees

Hai Duc Nguyen, Chaojie Zhang, Zhujun Xiao, and Andrew A Chien. Real-time serverless: Enabling application performance guarantees. In Proceedings of the 5th International Workshop on Serverless Computing, pages 1–6, 2019

work page 2019
[63]

Challenges in integration of heterogeneous internet of things.Scientific Program- ming, 2022(1):8626882, 2022

Muhammad Noaman, Muhammad Sohail Khan, Muhammad Faisal Abrar, Sikandar Ali, Atif Alvi, and Muhammad Asif Saleem. Challenges in integration of heterogeneous internet of things.Scientific Program- ming, 2022(1):8626882, 2022

work page 2022
[64]

In search of an understandable consensus algorithm

Diego Ongaro and John Ousterhout. In search of an understandable consensus algorithm. InProceedings of the USENIX annual technical conference (USENIX ATC 14), pages 305–319, 2014

work page 2014
[65]

Make applications faas- ready: Challenges and guidelines

Richard Patsch and Karl Michael G ¨oschka. Make applications faas- ready: Challenges and guidelines. InProceeding of the 6th International Conference on Information Technology and Computer Communications, pages 57–64, 2024

work page 2024
[66]

Scaling approaches for serverless data pipelines in edge and fog computing environments: A performance evaluation.ACM Transactions on Autonomous and Adaptive Systems, 2025

Shivananda Poojara, Pelle Jakovits, Rajkumar Buyya, and Satish Narayana Srirama. Scaling approaches for serverless data pipelines in edge and fog computing environments: A performance evaluation.ACM Transactions on Autonomous and Adaptive Systems, 2025

work page 2025
[67]

Chunkfunc: Dynamic slo-aware configuration of serverless functions.IEEE Transactions on Parallel and Distributed Systems, 2025

Thomas Pusztai and Stefan Nastic. Chunkfunc: Dynamic slo-aware configuration of serverless functions.IEEE Transactions on Parallel and Distributed Systems, 2025

work page 2025
[68]

Amir Masoud Rahmani, Amir Haider, Parisa Khoshvaght, Farhad Solei- manian Gharehchopogh, Komeil Moghaddasi, Shakiba Rajabi, and Mehdi Hosseinzadeh. Optimizing task offloading with metaheuristic algorithms across cloud, fog, and edge computing networks: A compre- hensive survey and state-of-the-art schemes.Sustainable Computing: Informatics and Systems, pa...

work page 2025
[69]

Serverless edge computing—where we are and what lies ahead.IEEE Internet Comput- ing, 27(3):50–64, 2023

Philipp Raith, Stefan Nastic, and Schahram Dustdar. Serverless edge computing—where we are and what lies ahead.IEEE Internet Comput- ing, 27(3):50–64, 2023

work page 2023
[70]

Nu: Achieving{Microsecond-Scale}resource fungibility with logical processes

Zhenyuan Ruan, Seo Jin Park, Marcos K Aguilera, Adam Belay, and Malte Schwarzkopf. Nu: Achieving{Microsecond-Scale}resource fungibility with logical processes. InProceedings of the 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23), pages 1409–1427, 2023

work page 2023
[71]

Serverless functions in the cloud-edge continuum: Challenges and oppor- tunities

Gabriele Russo Russo, Valeria Cardellini, and Francesco Lo Presti. Serverless functions in the cloud-edge continuum: Challenges and oppor- tunities. InProceedings of the 31st Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), pages 321–328. IEEE, 2023

work page 2023
[72]

Tempos: Qos management middleware for edge cloud computing faas in the internet of things.IEEE Access, 11, 2023

Gabriele Russo Russo et al. Tempos: Qos management middleware for edge cloud computing faas in the internet of things.IEEE Access, 11, 2023

work page 2023
[73]

Qos-aware offloading policies for serverless functions in the cloud-to-edge continuum.Future Generation Computer Systems, 156:1–15, 2024

Gabriele Russo Russo, Daniele Ferrarelli, Diana Pasquali, Valeria Cardellini, and Francesco Lo Presti. Qos-aware offloading policies for serverless functions in the cloud-to-edge continuum.Future Generation Computer Systems, 156:1–15, 2024

work page 2024
[74]

Towards qos-aware serverless function offloading in the edge-cloud continuum through reinforcement learning

Gabriele Russo Russo, Pierpaolo Spaziani, and Valeria Cardellini. Towards qos-aware serverless function offloading in the edge-cloud continuum through reinforcement learning. InProceedings of the IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pages 1073–1080. IEEE, 2025

work page 2025
[75]

Automated application deployment on multi-access edge computing: A survey.IEEE Access, 2023

´Alvaro Santos, Jorge Bernardino, and No ´elia Correia. Automated application deployment on multi-access edge computing: A survey.IEEE Access, 2023

work page 2023
[76]

Conflict-free replicated data types

Marc Shapiro, Nuno Preguic ¸a, Carlos Baquero, and Marek Zawirski. Conflict-free replicated data types. InStabilization, Safety, and Secu- rity of Distributed Systems: 13th International Symposium, SSS 2011, Grenoble, France, October 10-12, 2011. Proceedings 13, pages 386–

work page 2011
[77]

Video surveillance in smart cities: current status, challenges & future directions.Multimedia Tools and Applications, 84(16):15787–15832, 2025

Himani Sharma and Navdeep Kanwal. Video surveillance in smart cities: current status, challenges & future directions.Multimedia Tools and Applications, 84(16):15787–15832, 2025

work page 2025
[78]

Crdts as replication strategy in large-scale edge distributed system: An overview, 2020

Milo ˇs Simi ´c, Milan Stojkov, Goran Sladi ´c, and Branko Milosavljevi ´c. Crdts as replication strategy in large-scale edge distributed system: An overview, 2020

work page 2020
[79]

A survey of actor- like programming models for serverless computing

Jonas Spenger, Paris Carbone, and Philipp Haller. A survey of actor- like programming models for serverless computing. InActive Object Languages: Current Research Trends, pages 123–146. Springer, 2024

work page 2024
[80]

RKE2: Rancher’s Enterprise-Ready Kubernetes Distribution

SUSE Rancher Prime. RKE2: Rancher’s Enterprise-Ready Kubernetes Distribution. https://docs.rke2.io/. Online; Accessed on 8 Oct. 2025

work page 2025

Showing first 80 references.