Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Resource Allocation of Industry 4.0 Micro-Service Applications across Serverless Fog Federation

Razin Farhan Hussain, Mohsen Amini Salehi

TL;DR

This work tackles latency-constrained Industry 4.0 workloads at remote sites by introducing a serverless fog federation that dynamically coordinates nearby fogs. It advances two core components: ProPart, a probabilistic partitioning method for micro-service DAGs that maximizes the chance of on-time completion, and MR, a resource allocation strategy that selects execution sites based on end-to-end latency distributions and uncertainty, using ETC/ETT latency matrices and confidence-interval checks. Empirical results in EdgeCloudSim show that ProPart+MR achieve notable improvements in deadline adherence (roughly 15–20% under oversubscription) and makespan reductions, with enhanced fault tolerance as the federation scales. The approach enables cloud-like elasticity at the edge while remaining robust to network heterogeneity and operational outages, offering practical benefits for offshore and other remote industrial environments.

Abstract

The Industry 4.0 revolution has been made possible via AI-based applications (e.g., for automation and maintenance) deployed on the serverless edge (aka fog) computing platforms at the industrial sites -- where the data is generated. Nevertheless, fulfilling the fault-intolerant and real-time constraints of Industry 4.0 applications on resource-limited fog systems in remote industrial sites (e.g., offshore oil fields) that are uncertain, disaster-prone, and have no cloud access is challenging. It is this challenge that our research aims at addressing. We consider the inelastic nature of the fog systems, software architecture of the industrial applications (micro-service-based versus monolithic), and scarcity of human experts in remote sites. To enable cloud-like elasticity, our approach is to dynamically and seamlessly (i.e., without human intervention) federate nearby fog systems. Then, we develop serverless resource allocation solutions that are cognizant of the applications' software architecture, their latency requirements, and distributed nature of the underlying infrastructure. We propose methods to seamlessly and optimally partition micro-service-based application across the federated fog. Our experimental evaluation express that not only the elasticity is overcome in a serverless manner, but also our developed application partitioning method can serve around 20% more tasks on-time than the existing methods in the literature.

Resource Allocation of Industry 4.0 Micro-Service Applications across Serverless Fog Federation

TL;DR

This work tackles latency-constrained Industry 4.0 workloads at remote sites by introducing a serverless fog federation that dynamically coordinates nearby fogs. It advances two core components: ProPart, a probabilistic partitioning method for micro-service DAGs that maximizes the chance of on-time completion, and MR, a resource allocation strategy that selects execution sites based on end-to-end latency distributions and uncertainty, using ETC/ETT latency matrices and confidence-interval checks. Empirical results in EdgeCloudSim show that ProPart+MR achieve notable improvements in deadline adherence (roughly 15–20% under oversubscription) and makespan reductions, with enhanced fault tolerance as the federation scales. The approach enables cloud-like elasticity at the edge while remaining robust to network heterogeneity and operational outages, offering practical benefits for offshore and other remote industrial environments.

Abstract

The Industry 4.0 revolution has been made possible via AI-based applications (e.g., for automation and maintenance) deployed on the serverless edge (aka fog) computing platforms at the industrial sites -- where the data is generated. Nevertheless, fulfilling the fault-intolerant and real-time constraints of Industry 4.0 applications on resource-limited fog systems in remote industrial sites (e.g., offshore oil fields) that are uncertain, disaster-prone, and have no cloud access is challenging. It is this challenge that our research aims at addressing. We consider the inelastic nature of the fog systems, software architecture of the industrial applications (micro-service-based versus monolithic), and scarcity of human experts in remote sites. To enable cloud-like elasticity, our approach is to dynamically and seamlessly (i.e., without human intervention) federate nearby fog systems. Then, we develop serverless resource allocation solutions that are cognizant of the applications' software architecture, their latency requirements, and distributed nature of the underlying infrastructure. We propose methods to seamlessly and optimally partition micro-service-based application across the federated fog. Our experimental evaluation express that not only the elasticity is overcome in a serverless manner, but also our developed application partitioning method can serve around 20% more tasks on-time than the existing methods in the literature.
Paper Structure (21 sections, 2 equations, 12 figures, 2 tables, 1 algorithm)

This paper contains 21 sections, 2 equations, 12 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Overview and Motivation
  3. Problem Statement and Contributions
  4. Background and Prior Studies
  5. Resource Allocation in Industry 4.0
  6. Fog Computing for Industry 4.0 Use Cases
  7. ML Applications and Serverless Platform
  8. Software-hardware System Stack
  9. System Model
  10. Partitioning Micro-service Application Workflows
  11. Resource Allocation Method across Serverless Fog Federation
  12. Performance Evaluation
  13. Experimental Setup
  14. Baseline Workflow Partitioning Methods
  15. Baseline Resource Allocation Methods
  16. ...and 6 more sections

Figures (12)

  • Figure 1: Workflow of micro-services needed for the "fire detection" application. The application needs to seamlessly make use of federated fog to complete on time and prevent any potential damage.
  • Figure 2: A multi-layer view of the serverless fog federation infrastructure in the context of offshore smart oil and gas industry. Micro-service and monolithic Industry 4.0 applications seamlessly run across the federated fog systems by means of a platform, deployed on each gateway ($G_i$) representing a fog system.
  • Figure 3: System model of the proposed solution to allocate micro-service workflow or monolithic applications across a federation of three fog systems. Step 1 shows different applications arriving to the gateway of their local fog system. Step 2 shows the internal mechanics of each gateway that includes a "graph partitioning" and a "resource allocator" modules. The former is in charge of transparently partitioning the micro-service workflows to maximize its likelihood of on-time completion. The latter is in charge of seamlessly allocating each partition to a fog across the federation. Step 3 shows the serverless fog federation where each fog is represented by a gateway ($G_1, G_2, G_3$).
  • Figure 4: Flowchart of the ProPart workflow partitioning method. The output of this method is a partitioned workflow that is submitted to the resource allocation module, which is shown as the end step (with dashed lines) in this flowchart.
  • Figure 5: Comparison of the workflow partitioning methods. The horizontal axis represents the number of arriving workflow requests and the vertical axis shows the requests' deadline meeting rate.
  • ...and 7 more figures