Source-linked AI summary
FogBus: A Blockchain-based Lightweight Framework for Edge and Fog Computing
Shreshth Tuli, Redowan Mahmud, Shikhar Tuli, Rajkumar Buyya
TL;DR
FogBus targets limitations in integrating heterogeneous IoT, Fog, and Cloud infrastructure for latency-sensitive and computing-intensive applications. It provides a lightweight, platform-independent PaaS framework with multi-application execution and Blockchain-based security, and its Sleep Apnea prototype evaluation examines latency, energy, network, and CPU usage.
Problem
Existing IoT–Fog–Cloud frameworks have limitations in platform independence, security, resource management, and simultaneous multi-application support.
Method
FogBus integrates IoT, Fog, and Cloud resources through platform-independent execution and interaction interfaces, a PaaS model, and Blockchain, authentication, and encryption mechanisms.
Results
FogBus is evaluated in a Sleep Apnea analysis environment, with performance assessed across latency, energy, network, and CPU usage under different settings.
Takeaways & Limitations
The framework supports end-to-end IoT–Fog–Cloud deployment, application customization, resource management, and protected handling of sensitive data.
Takeaways & Limitations
FogBus’s existing security features require comparatively higher computational assistance and affect service delivery latency, energy, and network usage.
Abstract
from arXiv · showhide
The requirement of supporting both latency sensitive and computing intensive Internet of Things (IoT) applications is consistently boosting the necessity for integrating Edge, Fog and Cloud infrastructure. Although there are a number of real-world frameworks attempt to support such integration, they have many limitations from various perspectives including platform independence, security, resource management and multi-application assistance. To address these limitations, we propose a simplified but effective framework, named FogBus for facilitating end-to-end IoT-Fog(Edge)-Cloud integration. FogBus offers a platform independent interface to IoT applications and computing instances for execution and interaction. It not only assists developers in building applications but also helps users in running multiple applications at a time and service providers to manage their resources. In addition, FogBus applies Blockchain, authentication and encryption techniques to secure operations on sensitive data. Because of its lightweight and cross platform software systems, it is easy to deploy, scalable and cost e_cient. We demonstrate the effectiveness of our framework by creating a computing environment with it that integrates finger pulse oximeter as IoT devices with Smartphone-based gateway and Raspberry Pi-based Fog nodes for Sleep Apnea analysis. We also run several experiments on this computing environment varying FogBus settings. The experimental results show that different FogBus settings can improve latency, energy, network and CPU usage of the computing infrastructure.
1. Introduction
FogBus addresses limitations in integrating heterogeneous IoT, Fog, and Cloud resources for latency-sensitive and computing-intensive applications. It combines platform-independent execution, multi-application support, resource management, and security mechanisms, and evaluates the framework across several infrastructure metrics.
- Motivation: Fog and Edge computing bring processing closer to IoT data sources, but Fog resources remain heterogeneous and constrained.Cloud-centric processing can create congestion, while Fog cannot accommodate every compute-intensive application.
- Research Gap: Existing integration frameworks provide limited platform independence, simultaneous multi-application execution, customization, resource management, and security.They also rely heavily on Cloud storage, centralized management, or processing by energy-constrained IoT devices.
- Approach: FogBus integrates IoT systems with Fog and Cloud resources through a lightweight framework supporting platform-independent execution and node-to-node interaction.The framework is intended to harness edge and remote resources according to application requirements.
- Approach: FogBus provides a PaaS model for developers, users, and service providers while supporting simultaneous latency-sensitive and compute-intensive applications.Its functions include application development, service customization, and resource management.
- Security: Blockchain, authentication, and encryption are used to protect sensitive data and support integrity, protection, and privacy.The paper specifically identifies Blockchain as an integrity mechanism during data transfer.
- Evaluation: FogBus is evaluated using latency, energy, network, and CPU usage in a Sleep Apnea analysis prototype.The prototype integrates IoT devices, gateways, Fog nodes, and Cloud resources.
2. Related Work
Prior IoT–Fog–Cloud frameworks include application-specific prototypes and generalized PaaS models, but the literature identifies recurring gaps in security, resource utilization, heterogeneity, and multi-application support. FogBus combines prototype and platform-based ideas to address these gaps.
- Framework Categories: Existing frameworks are broadly classified as application-specific prototypes or generalized PaaS models.Table 1 summarizes the literature study.
- Application-Specific Prototypes: Prototype frameworks target domains including healthcare, smart-city surveillance, gas-leak monitoring, and face identification.Reported designs variously use gateways, Fog processing, Cloud backends, and local preprocessing.
- PaaS Frameworks: PaaS frameworks support application development, deployment, node management, heterogeneity handling, security features, or task migration with differing scopes.Examples include Cloud-centric, Fog-centric, gateway-based, and service-oriented designs.
- Identified Gaps: The reviewed frameworks often provide limited security and fail to fully leverage edge and remote computing resources.Pushing computation toward constrained devices or resource-rich Fog nodes can increase deployment cost and energy consumption.
- Identified Gaps: Most reviewed frameworks overlook infrastructure heterogeneity and have difficulty supporting multiple applications simultaneously.These limitations motivate FogBus’s combined prototype and platform-based design.
3. FogBus Framework
FogBus integrates diverse IoT, Fog, and Cloud hardware through platform-independent software components and coordinated services. Its architecture assigns gateway, computational, repository, security, resource, data, and Cloud-integration responsibilities across the environment.
- Architecture: FogBus uses structured communication and platform-independent execution interfaces to integrate diverse hardware instruments.The framework’s high-level architecture connects IoT, Fog, and Cloud components.
- Hardware Instruments: IoT devices produce or consume sensed data, Fog Gateway Nodes connect them to the infrastructure, and Fog Computational Nodes execute application operations.IoT devices may preprocess data, while gateways provide configuration and user interfaces.
- Computational Roles: Broker nodes coordinate back-end processing across Fog nodes and Cloud datacenters when local resources cannot satisfy application requirements.They distribute tasks, monitor activities, and synchronize and coordinate execution.
- System Services: FogBus organizes platform-independent software into Broker, Computing, and Repository services.Broker manages broker-node functionality, Computing controls general nodes, and Repository supports data and application information.
- Security and Data: Security components authenticate users, manage keys and certificates, create Blockchain records, and protect communication and stored data.The design combines credential archives, security managers, encryption, and Blockchain-related verification.
- Resource Management: Resource Manager selects execution resources using application requirements and node status, with Cloud Integrator supplying Cloud-instance context.Application Catalogue records application operations, dependencies, and resource requirements.
4. Design and Implementation
FogBus implements platform-independent software services for coordinating applications, resources, storage, security, and Cloud integration across heterogeneous IoT-Fog-Cloud infrastructure.
- System Services: FogBus uses platform-independent APIs, execution environments, scripting, and programming languages to operate across heterogeneous hardware.These capabilities are supported across the integrated environment, helping FogBus function beyond infrastructure heterogeneity.
- System Services: Broker, Repository, and Computing services coordinate node interaction, resource management, data storage, and application execution.The services are implemented as web programs using PHP, HTTP, and RESTful APIs for exchanging information among Fog computing nodes.
- Blockchain: Blockchain protection combines SHA256-linked blocks, nonce values, timestamps, digital signatures, and public/private keys for data integrity and source verification.Masters create blocks from received data and share Blockchain details and signature attributes with workers.
- Cloud Plugin: FogBus activates its Cloud Plugin for computation only when users request Cloud resources, while other Cloud operations remain managed directly.The current implementation uses Aneka as the third-party software for Cloud computational integration.
- Resource Provisioning: Fog resources are used before Cloud resources under FogBus’s built-in provisioning policy.When Cloud processing is needed, the Aneka-based plugin polls the Cloud input file every 500 milliseconds for pending data.
- Application Development: Applications separate user interfaces running on Fog gateway nodes from distributed backend programs executed across computing nodes.Developers are advised to use platform-compatible interface technologies and cross-platform, modular backend programs that handle intermediate and output files.
5. A Case Study : Sleep Apnea Analysis
The case study deploys FogBus for low-cost, latency-sensitive Sleep Apnea analysis using a pulse oximeter, smartphone gateway, Fog nodes, and optional Cloud processing.
- Motivation: Sleep Apnea interrupts airflow for at least 10 seconds during sleep and can reduce blood oxygen and heartbeat rate.Monitoring blood oxygen saturation over time helps determine disease intensity and identify when medical consultation is recommended.
- Motivation: The prototype gathers SpO2 and heart beat rate from a finger pulse oximeter and harnesses local resources for processing.It is intended to provide an affordable, configurable, and faster alternative to expensive hospital or laboratory-based analysis.
- System Setup: The implementation combines a Bluetooth Low Energy finger pulse oximeter, Android smartphone gateway, Dell master node, Raspberry Pi worker node, and Microsoft Azure Cloud.The listed devices and software form the prototype’s IoT-Fog-Cloud environment.
- System Architecture: FogBus primarily processes Sleep Apnea data on Fog infrastructure and sends it to an Azure virtual machine when Fog resources cannot process it.The fallback uses FogBus’s built-in Aneka-based Cloud Plugin.
- Installed Package: The HealthKeeper Android interface pairs the oximeter, accepts the master’s IP address, and manages data transmission through Home and Session screens.The Home screen handles pairing and configuration, while the Session screen handles interaction with the master.
- Data Analytic: The worker analytic detects oxygen dips below 88, verifies them using nearby heart-rate rises, and counts confirmed dips as AHI.The analytic also reports oxygen and heart-rate statistics, filters heartbeat patterns during dips, and generates ECG output.
- Communication Sequence: The communication sequence sends oximeter measurements through the smartphone to the master, which stores data and later assigns analysis to a computing worker.The worker retrieves stored data and the analytic executable before starting analysis.
6. Performance Evaluation
FogBus performance was evaluated using a Sleep Apnea analysis prototype across settings that varied task intervals, Fog/Cloud placement, and Blockchain use. Fog-only and disabled-Blockchain configurations generally reduced latency, network, energy, and master-resource usage, while task intervals traded throughput for lower overhead.
- Experimental Setup: The Sleep Apnea prototype evaluated latency, energy, CPU, RAM, Cache, and network usage using pulse-oximeter data and sequential analysis tasks.Experiments were monitored at the master, Azure VM, and Raspberry Pi circuits.
- Number of Tasks: Fog Only generated more tasks than Cloud Only and Integrated Fog-Cloud, while Without Interval generated more tasks than With Interval.Fog infrastructure delivered previous outcomes quickly, and continuous task generation increased task production.
- Latency: Fog Only minimized service delivery latency, with further reductions when Blockchain was disabled and task intervals reduced infrastructure and network overhead.The experiment attributed the latency advantage mainly to lower network propagation delay because task data chunks were not huge.
- Network Usage: Fog Only reduced network usage relative to Cloud Only and Integrated Fog-Cloud, while disabling Blockchain reduced transferred security attributes.Continuous task generation increased network usage, which could be reduced by tuning the interval between tasks.
- Energy: Fog Only required less energy than Cloud Only, and disabling Blockchain plus task intervals further reduced energy consumption.Cloud VMs consumed more energy than Fog nodes, while longer intervals reduced busy-time consumption but also processed fewer tasks.
- CPU, RAM, Cache Usage of Broker / Master: Fog Only used less master CPU, RAM, and Cache than Cloud Only, while disabling Blockchain reduced CPU-related overhead; RAM and Cache remained nearly unchanged across settings.The paper attributes persistent RAM and Cache usage to FogBus components not releasing allocated resources after operations.
7. Conclusion and Future Works
FogBus is a lightweight, cross-platform framework for integrating IoT systems with Fog and Cloud infrastructure while supporting application development, service customization, resource management, and secure data handling. Future work targets dynamic resource management, Fog virtualization, artificial intelligence, application placement and migration, and lighter security features.
- FogBus integrates IoT-enabled systems with Fog and Cloud infrastructure using lightweight, cross-platform software.The framework is intended to harness both edge and remote resources for application deployment, monitoring, and management.
- FogBus operates as a PaaS model supporting application development, service customization, and context-aware resource management.
- Authentication, Blockchain, and encryption address data privacy, integrity, and transfer across less secure networks.
- Resource management policies: Dynamic resource management policies remain future work for load balancing and QoS enhancement beyond the existing static policy.
- Future works: Future extensions include Fog infrastructure virtualization, artificial intelligence, application placement, runtime migration, and lightweight security features.The paper specifically notes that Fog virtualization needs deeper exploration, while current security features require comparatively higher computational assistance.