Modeling Availability in Softwarized MEC: Integrating a Fault Injection Tool for Effective Validation

  • Erick Nascimento UFPE
  • Eduardo Antonio Guimaraes Tavares UFPE
  • Jamilson Dantas UFPE
  • Paulo Maciel UFPE

Resumo


Fifth-generation (5G) networks promise ultra-fast speeds, minimal latency, and improved reliability, essential for supporting critical applications such as healthcare systems. Stochastic models are prominent in addressing the Multi-access edge computing (MEC) system’s complexities and evaluating dependability metrics, such as availability. This strategy enables infrastructure planners to make informed decisions, improving reliability and performance in SDN-based networks. This paper proposed a hierarchical modeling evaluation and validation through continuous-time Markov chains (CTMC) and reliability block diagrams (RBD) to analyze the availability of an SDN-based MEC system. A software-based fault injection tool was adopted to transmit faults and repairs on the testbed to verify a system’s availability by ensuring it could handle faults without fatal failures. This method is essential for systems that require high availability and reliability, such as critical infrastructure, medical devices, and aerospace systems. Next, two case studies demonstrate the practical application of the models and the use of the fault injection tool to validate the system practically. It also outlines future research directions in SDN infrastructure for MEC systems.

Palavras-chave: Availability evaluation, Fault injection, Monitoring, Hierarchical models, Multi-access edge computing, Software-defined networks
Publicado
26/11/2024
NASCIMENTO, Erick; TAVARES, Eduardo Antonio Guimaraes; DANTAS, Jamilson; MACIEL, Paulo. Modeling Availability in Softwarized MEC: Integrating a Fault Injection Tool for Effective Validation. In: LATIN-AMERICAN SYMPOSIUM ON DEPENDABLE COMPUTING (LADC), 13. , 2024, Recife/PE. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 38–48.