Source-linked AI summary
Transformative effects of IoT, Blockchain and Artificial Intelligence on cloud computing: Evolution, vision, trends and open challenges
Sukhpal Singh Gill, Shreshth Tuli, Minxian Xu, Inderpreet Singh, Karan Vijay Singh, Dominic Lindsay, Shikhar Tuli, Daria Smirnova, Manmeet Singh, Udit Jain, Haris Pervaiz, Bhanu Sehgal, Sukhwinder Singh Kaila, Sanjay Misra, Mohammad Sadegh Aslanpour, Harshit Mehta, Vlado Stankovski, Peter Garraghan
TL;DR
Cloud systems must handle growing demand and heterogeneous QoS requirements while addressing energy, reliability, security, and resource-management challenges. The paper systematically reviews computing evolution and the influence of IoT, AI, and Blockchain, incorporates expert perspectives, and proposes a conceptual model for cloud futurology. It concludes that these paradigms define important research directions, while their integration remains constrained by resource, scalability, privacy, fault-tolerance, and data-analytics challenges.
Problem
Cloud computing must satisfy diverse QoS requirements under growing demand and data volumes while addressing energy, reliability, security, and resource-management challenges.
Method
The paper systematically reviews computing paradigms, technologies, and research challenges, incorporates international expert perspectives, and proposes a conceptual model for cloud evolution.
Results
The review identifies the influence of IoT, AI, and Blockchain on cloud evolution and synthesizes application directions including healthcare, retail, analytics, and cryptocurrency.
Takeaways & Limitations
The paper frames integrated IoT, AI, and Blockchain as key directions for next-generation cloud computing and highlights associated open challenges.
Takeaways & Limitations
Blockchain integration in IoT and fog environments remains limited by resource constraints, overheads, privacy, scalability, fault rates, and single points of failure.
Abstract
from arXiv · showhide
Cloud computing plays a critical role in modern society and enables a range of applications from infrastructure to social media. Such system must cope with varying load and evolving usage reflecting societies interaction and dependency on automated computing systems whilst satisfying Quality of Service (QoS) guarantees. Enabling these systems are a cohort of conceptual technologies, synthesized to meet demand of evolving computing applications. In order to understand current and future challenges of such system, there is a need to identify key technologies enabling future applications. In this study, we aim to explore how three emerging paradigms (Blockchain, IoT and Artificial Intelligence) will influence future cloud computing systems. Further, we identify several technologies driving these paradigms and invite international experts to discuss the current status and future directions of cloud computing. Finally, we proposed a conceptual model for cloud futurology to explore the influence of emerging paradigms and technologies on evolution of cloud computing.
1. Introduction
Cloud computing has evolved into a widely adopted utility that must provide heterogeneous QoS under changing demand. This review examines how IoT, AI, and Blockchain may shape its future and proposes a conceptual model for cloud futurology.
- Cloud computing emerged from distributed-systems research and provides dynamic, metered access to shared computing resources.
- Large-scale cloud data centers provision heterogeneous QoS requirements through unified interfaces over Internet resources.
- IoT-based applications can improve cloud-service reliability, while SLAs specify service delivery time, budget, and QoS parameters.
- Growing demand and data volumes increase data-center energy consumption, emissions, and cooling needs, motivating resource scheduling that preserves QoS.
- The review revisits cloud evolution through IoT, AI, and Blockchain, synthesizes expert perspectives, and identifies future research directions.
- A conceptual model is proposed to integrate emerging technologies and explore their influence on next-generation cloud computing.
2. Background: History of Decades
Computing paradigms evolved from single-task systems toward interconnected and parallel architectures, progressively improving resource use and execution capacity. The background traces this history through supercomputing, mainframes, clusters, parallel processing, fog, and IoT.
- Computing systems evolved from 1958 onward through paradigms and technologies intended to use hardware resources more efficiently.
- Early systems required multiple computers to execute multiple tasks concurrently, motivating secure communication networks for data exchange.
- Supercomputers provide high-performance execution for computationally intensive scientific tasks, while energy use and heat management remained persistent challenges.
- Cluster computing coordinates interconnected nodes over fast local networks to execute tasks and improve computing performance.
- Fog computing supports IoT devices including mobile phones, sensors, and health-monitoring devices.
3. Triumvirate: IoT + AI + Blockchain
IoT, Blockchain, and AI extend cloud computing toward intelligent, distributed, and more secure systems. Their integration offers lower-latency processing, data integrity, and adaptive optimization, but remains constrained by privacy, scalability, resources, and fault tolerance.
- IoT: IoT connects sensors, actuators, appliances, and software across heterogeneous application domains for monitoring, control, and data exchange.
- IoT: Cloud offloading struggles with IoT real-time demands because geographically distributed devices generate high-density data requiring low latency and bandwidth.
- IoT: Fog and edge computing decentralize orchestration toward local nodes, reducing response latency, increasing ingestion bandwidth, and supporting mobility.
- IoT: Smart e-health systems process wearable and implantable-device data locally while using cloud layers for long-term trend analysis.
- Blockchain: Blockchain records data in distributed ledgers and uses proof-of-work validation to make timely fraudulent alteration difficult.
- Blockchain: Fog–Blockchain frameworks can dynamically allocate less-utilized nodes for mining and validation while other nodes handle computation and data collection.
- Blockchain: Blockchain integration faces storage, scalability, privacy, resource, bandwidth, redundancy, and consensus challenges in heterogeneous fog environments.
- Artificial Intelligence: AI techniques help IoT and fog nodes adapt to workload conditions while targeting improved QoS, lower power consumption, or reduced infrastructure cost.
4. Impact of New Paradigms and Technologies on Cloud Computing: Open Challenges and Trends
The section surveys emerging cloud-computing research areas and identifies open challenges involving QoS, fog computing, energy management, and resource-intensive IoT deployments.
- 4.1 QoS and SLA: QoS research must address resource unavailability, SLA-aware management, differing provider standards, and trade-offs among requirements in IoT applications.AI-based learning and prediction models are identified as possible approaches for managing these trade-offs.
- 4.2 Fog Computing: Fog computing places an intermediate layer between IoT devices and cloud data centers, reducing latency for time-sensitive applications while remaining unable to replace cloud processing of heavy data.Fog also provides mobility support, security, performance, and lower bandwidth requirements.
- 4.2 Fog Computing: Fog and cloud scheduling remains difficult because heterogeneous resources and changing environments require optimization across multiple objectives rather than isolated energy, response-time, or SLA metrics.Existing heuristics and adaptive schedulers do not fully optimize diverse user or application requirements.
- Open Challenges and Trends: Future directions include generic fog interfaces, blockchain APIs suitable for resource-constrained edge nodes, and AI-based task scheduling for heterogeneous fog environments.The section also calls for blockchain designs that support energy-constrained execution and AI techniques for thermal-aware scheduling.
- 4.3 Energy Management: Cloud energy consumption is rising with data-center growth, motivating distributed Fog and Edge platforms and efficiency techniques across software, hardware, bandwidth, thermal scheduling, and energy harvesting.The proposed directions include more compact data encoding, thermal-aware scheduling, and reduced-power nodes.
4.4 Resource Management
Resource management must coordinate heterogeneous distributed infrastructure while preserving performance, latency sensitivity, reliability, and energy efficiency. The section proposes AI, blockchain, and fault-tolerance directions for improving allocation, scheduling, and service continuity.
- Resource Management: Resource management becomes more complex as distributed systems scale across diverse network devices, components, communication technologies, servers, and cooling infrastructure.The required systems must provide high performance, latency sensitivity, reliability, and energy efficiency.
- Resource Management: AI techniques and experiment-driven approaches are proposed for holistic resource management, with data abstraction illustrated by cluster-management systems such as Borg.Borg hides resource-management details so users can focus on applications.
- Resource Management: Mesos supports fair sharing of commodity clusters by making resource offers according to framework constraints while leaving offer acceptance and scheduling decisions to frameworks.This design distributes scheduling responsibility between Mesos and the frameworks running on the cluster.
- Open Challenges and Trends: Future resource-management work should provide QoS-aware autonomic management, balanced blockchain mining and hash allocation, and AI-based prediction for geographically distributed resources.The proposed scheduling policies aim to predict resource requirements in advance and avoid runtime SLA violations.
- 4.5 Fault Tolerance: Fault-tolerance research addresses continuous reliable service through failure prediction, replication for IoT tasks, and reliable storage for big-data processing.Replication-based techniques are described as improving task latency and response time.
- 4.6 Security and Privacy: Security and privacy models face scalability limits in next-generation edge computing, while decentralized fog layers introduce threats not present in conventional cloud computing.Future work includes broader edge-resource use, AI-based attack mitigation, and security mechanisms accommodating mobile devices.
4.7 Software-Defined Network
Software-defined networking and related analytics technologies extend cloud resource virtualization while addressing flexibility, cost, energy, security, and reliability challenges. The section also connects cloud-based analytics with IoT, healthcare, government, and retail applications.
- SDN virtualizes network, storage, and compute resources to improve abstraction, automation, configuration optimization, and network agility.
- NFV executes functions such as DNS, load balancing, and intrusion detection as software, improving network elasticity and flexibility while reducing cost.
- Open Challenges and Trends: SDN-based cloud computing still requires stronger security for inter-datacenter data transfer, blockchain-oriented decentralization, and AI-based deployment for lower energy use and higher reliability.
- Applications: Analytics applications include healthcare services that support earlier disease detection and customized treatment, and government planning and decision-making based on data patterns and sentiments.
- Applications: Retail analytics correlates purchase histories, profiles, and social-media behavior to reveal customer insights, support timely offers, and predict churn.
- Big Data Analytics and Data Science: Cloud-based big-data analytics uses machine learning and pattern recognition to uncover sensor and machine-data insights, predict trends and outliers, and improve operational decisions.
4.9 Data Processing
Data processing is increasingly central to cloud and IoT systems because rapidly growing, heterogeneous data must be managed, analyzed, and protected at scale. The section emphasizes latency, scalability, privacy, and reliability requirements, with fog and edge computing as supporting approaches.
- Data processing supports analytics by collecting and manipulating information into usable forms, but consumes nearly 70-80% of data analysts’ and scientists’ project time.
- Batch, real-time, and hybrid processing address different combinations of data volume, velocity, latency, and security requirements.
- Cloud adoption helps organizations scale data storage, computation, and analysis as on-premises advanced-analytics infrastructure becomes expensive and demanding.
- IoT devices and applications generate rapidly expanding data that requires scalable, reliable, fast, private, and secure cloud-supported architectures.The passage estimates 50 billion online devices and 40% of world data originating from them by 2020, with $1.7 trillion in expenditure.
- Open Challenges and Trends: Application design must address latency, bandwidth, storage capacity, maintenance cost, scalability, and blockchain-security overhead in IoT cloud systems.
- Fog Computing: Fog computing stores and processes some IoT data locally, reducing excessive cloud-resource use and supporting reliable responses for time-sensitive, data-intensive applications.
- Applications: IoT, big-data analytics, fog, and edge computing support smart-city, transportation, agriculture, and healthcare applications.
4.11 Serverless Computing
Serverless and containerized cloud applications shift execution and resource management toward managed or lightweight environments, but introduce QoS, security, privacy, and resource-constraint challenges. The section identifies blockchain and AI as possible directions for addressing these issues.
- Serverless applications place database and application logic in the cloud while forwarding business logic to end users for execution on provisioned resources without renting virtual machines.
- Open Challenges and Trends: Serverless scheduling must account for cold-start delays, shared-file bandwidth in Bag-of-Tasks workloads, and application and data security and privacy.
- Edge devices integrated into serverless datacenters are resource-constrained and cannot support heavy security applications and firewalls designed for personal computers.
- Open Challenges and Trends: Proposed directions include blockchain for edge-device security, new IoT applications for secure communication and privacy, and AI systems for serverless application design.
- Deep Learning: Distributed deep-neural-network training across edge and IoT devices can reduce communication cost by 20x, while mobile-cloud partitioning can reduce neural-network feature size and energy consumption.
- Deep Learning: A hybrid deep-learning model operating partly on IoT devices and partly in the cloud achieved 95.84% accuracy with robustness across varied datasets.
- Cloud Containers: Container technologies provide lightweight, isolated environments with resource sharing, but kernel sharing creates security weaknesses compared with virtual machines.
- Cloud Containers: Future container research includes Unikernel-based security, performance optimization, blockchain adaptation, and machine-learning management of QoS-based clusters.
4.14 Quantum Computing
The section surveys quantum computing’s potential applications in optimization, machine learning, material simulation, and blockchain-related security, while identifying limitations of blockchain resources and latency.
- Quantum computing uses superposition and entanglement to address problems that classical computers cannot solve at certain sizes and complexities.
- Optimization: Quantum computing is discussed for optimization problems such as dynamic asset allocation, airline traffic scheduling, financial analysis, and system design.
- Machine Learning and AI: Quantum computing may transform machine learning, including object detection trained on large datasets.
- Material Simulation: Material simulation could support IoT applications in robotic, chemical, and optical industries.
- Blockchain and Cloud: Blockchain-based distributed clouds are proposed to combine blockchain features with cloud capacity while addressing blockchain’s limited storage, inefficient virtual machines, and high latency.
4.17 5G and Beyond
The section presents 5G, edge computing, AI, and blockchain as components of future cloud and IoT systems, while emphasizing unresolved latency, security, prediction, and coordination challenges.
- 5G and Beyond: 5G is expected to provide high-speed data rates, reduced latency, reduced energy consumption, ubiquitous coverage, and massive IoT connectivity.
- Edge Computing: Edge computing shifts application hosting from centralized data centres toward the network edge, closer to consumers and generated data.
- 5G and Beyond: The advent of 5G and cloud computing is described as enhancing network operators’ capacity, functionality, and flexibility to offer new services.
- Open Challenges and Trends: Real-time communication systems face bottlenecks from computation-intensive IoT applications and growing mission-critical workloads in 5G networks.
- Autoscaling: Fog and edge environments require collaboration among clouds and proximity to users, but raise monetization concerns and depend on 5G networks still in early development.
- Autoscaling: Future autoscaling research includes compute-continuum scaling, blockchain-enabled monetization and security, and AI-based self-correction with multi-objective performance–cost trade-offs.
5. Insights of Triumvirate to the Cloud Computing Evolution: A vision
Figure 3 presents insights concerning the triumvirate’s relationship to cloud computing evolution.
- Figure 3 is titled “Insights of Triumvirate to the Cloud Computing Evolution.”
- The figure concerns cloud computing evolution.
- The figure identifies the triumvirate as its subject.
6. A Conceptual model for Cloud Futurology: Holistic View
The conceptual model integrates Blockchain, IoT, and Artificial Intelligence to examine their transformation effects on cloud computing evolution from a holistic perspective.
- The model is proposed to explore how Blockchain, IoT, and Artificial Intelligence influence cloud computing evolution.
- Figure 4 presents a conceptual model describing the transformation effects of paradigms and technologies on cloud computing evolution.
- The model integrates the three paradigms into a holistic view of an abstract design.
7. Summary and Conclusions
The paper reviews cloud computing paradigms and technologies and examines how Blockchain, IoT, and Artificial Intelligence influence cloud-computing evolution. It presents a conceptual model for Cloud Futurology.
- Cloud computing provides on-demand, metered access to compute resources, including processing, memory, and storage.
- The review examines computing paradigms and technologies and the influence of Blockchain, IoT, and Artificial Intelligence on cloud-computing evolution.
- The paper presents a conceptual model for Cloud Futurology.
Declaration of Competing of Interest
The article declares that it has no conflicts of interest.
- The authors declare no conflicts of interest.