Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Self-Healing and Fault-Tolerant Cloud-based Digital Twin Processing Management Model

Deepika Saxena, Ashutosh Kumar Singh

TL;DR

Cloud-based digital twins face outages, security threats, and resource contention that hinder reliable DT development. The authors propose SF-DTM, a Self-Healing and Fault-tolerant cloud-based Digital Twin Processing Management model that combines SimiFed (cosine-similarity guided Federated Learning with LSTM) for collaborative processing estimation, a self-healing workflow with frequent sequence pattern analytics, and MVP-based replication for fault tolerance. Empirical evaluation on real workload traces shows improved availability, MTBF, and MTTR compared with non-SF-DTM baselines, including up to 13.2% higher service availability. The approach preserves data privacy, reduces downtime, and scales with increasing DT task sizes, offering a robust framework for reliable, collaborative DT management in multi-tenant cloud environments.

Abstract

Digital twins, integral to cloud platforms, bridge physical and virtual worlds, fostering collaboration among stakeholders in manufacturing and processing. However, the cloud platforms face challenges like service outages, vulnerabilities, and resource contention, hindering critical digital twin application development. The existing research works have limited focus on reliability and fault tolerance in digital twin processing. In this context, this paper proposed a novel Self-healing and Faulttolerant cloud-based Digital Twin processing Management (SF-DTM) model. It employs collaborative digital twin tasks resource requirement estimation unit which utilizes newly devised Federated learning with cosine Similarity integration (SimiFed). Further, SF-DTM incorporates a self-healing fault-tolerance strategy employing a frequent sequence fault-prone pattern analytics unit for deciding the most admissible VM allocation. The implementation and evaluation of SF-DTM model using real traces demonstrates its effectiveness and resilience, revealing improved availability, higher Mean Time Between Failure (MTBF), and lower Mean Time To Repair (MTTR) compared with non-SF-DTM approaches, enhancing collaborative DT application management. SF-DTM improved the services availability up to 13.2% over non-SF-DTM-based DT processing.

A Self-Healing and Fault-Tolerant Cloud-based Digital Twin Processing Management Model

TL;DR

Cloud-based digital twins face outages, security threats, and resource contention that hinder reliable DT development. The authors propose SF-DTM, a Self-Healing and Fault-tolerant cloud-based Digital Twin Processing Management model that combines SimiFed (cosine-similarity guided Federated Learning with LSTM) for collaborative processing estimation, a self-healing workflow with frequent sequence pattern analytics, and MVP-based replication for fault tolerance. Empirical evaluation on real workload traces shows improved availability, MTBF, and MTTR compared with non-SF-DTM baselines, including up to 13.2% higher service availability. The approach preserves data privacy, reduces downtime, and scales with increasing DT task sizes, offering a robust framework for reliable, collaborative DT management in multi-tenant cloud environments.

Abstract

Digital twins, integral to cloud platforms, bridge physical and virtual worlds, fostering collaboration among stakeholders in manufacturing and processing. However, the cloud platforms face challenges like service outages, vulnerabilities, and resource contention, hindering critical digital twin application development. The existing research works have limited focus on reliability and fault tolerance in digital twin processing. In this context, this paper proposed a novel Self-healing and Faulttolerant cloud-based Digital Twin processing Management (SF-DTM) model. It employs collaborative digital twin tasks resource requirement estimation unit which utilizes newly devised Federated learning with cosine Similarity integration (SimiFed). Further, SF-DTM incorporates a self-healing fault-tolerance strategy employing a frequent sequence fault-prone pattern analytics unit for deciding the most admissible VM allocation. The implementation and evaluation of SF-DTM model using real traces demonstrates its effectiveness and resilience, revealing improved availability, higher Mean Time Between Failure (MTBF), and lower Mean Time To Repair (MTTR) compared with non-SF-DTM approaches, enhancing collaborative DT application management. SF-DTM improved the services availability up to 13.2% over non-SF-DTM-based DT processing.
Paper Structure (17 sections, 10 equations, 10 figures, 4 tables, 1 algorithm)

This paper contains 17 sections, 10 equations, 10 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Paper Contributions
  4. Paper Organization
  5. Proposed Model
  6. SimiFed: DT application processing estimation
  7. Self-Healing and Fault-tolerant strategy
  8. Assignment and Execution
  9. Operational Design and Complexity
  10. Performance Evaluation
  11. Experimental set-up and Dataset
  12. Key Performance Indicators
  13. Results and Discussion
  14. DT processing requirement estimation metrics
  15. Fault pattern generation metrics
  16. ...and 2 more sections

Figures (10)

  • Figure 1: Cloud-driven collaborative DT and challenges
  • Figure 2: SF-DTM Model
  • Figure 3: Frequent sequence pattern analytics
  • Figure 4: Observed calibration over consecutive time duration of 300 minutes
  • Figure 5: DT application fault estimation metrics
  • ...and 5 more figures