Enhancing Scalability and Low Latency in Software Defined Fog of Things: A new Intelligent Scheduler and Load Balancer

Nome do aluno Ernando Passos Batista Junior
Título do trabalho Enhancing Scalability and Low Latency in Software Defined Fog of Things: A new Intelligent Scheduler and Load Balancer
Resumo do trabalho The significant adoption of the Internet of Things (IoT) has intensified the challenges in providing suitable IoT infrastructures that meet essential requirements, such as low latency and scalability. In dynamic environments where IoT devices frequently change communication behavior or are added to the network, ensuring timely responses and balanced resource utilization becomes increasingly complex. This thesis presents Software-Defined Fog of Things (SD-FoT). This architecture leverages Software-Defined Networking (SDN) and the OpenFlow protocol to enable programmable, intelligent management of Fog of Things (FoT) based infrastructures. SD-FoT incorporates two key components: the SD-FoT Scheduler proposal, which uses Long Short-Term Memory (LSTM) based time series forecasting to anticipate communication patterns and proactively install OpenFlow rules, reducing latency in communications in SD-FoT; and the SD-FoT Load Balancer proposal, which also uses machine learning to predict and mitigate overload scenarios among FoT nodes and network paths, ensuring system scalability and low latency. A Systematic Literature Review was also conducted to consolidate current approaches involving SDN and machine learning in IoT-Fog scenarios, identifying gaps addressed by this proposal. Experimental validation is supported by FoT-Simulator, a high-fidelity and custom-built emulation environment. Together, these contributions provide a scalable, low-latency solution for managing the increasing demands of modern IoT systems in Fog of Things contexts.
Orientador Prof. Dr. Cássio Prazeres (PGCOMP-UFBA)
Co-orientador Prof. Dr. Gustavo B. Figueiredo (PGCOMP-UFBA)
Membro externo 1 (com afiliação) Profa. Dra. Flávia Coimbra Delicato (UFF - PQ 1D)
Link para o curriculum lattes http://lattes.cnpq.br/5386282151810710
Membro interno 1 (com afiliação) Prof. Dr. Bruno Pereira dos Santos (PGCOMP-UFBA)
Link para o curriculum lattes http://lattes.cnpq.br/0092226104911153
Suplente do membro externo (com afiliação) Prof. Dr. Viní­cius Fernandes Soares Mota (UFES)
Link para o curriculum lattes http://lattes.cnpq.br/9305955394665920
Suplente do membro interno (com afiliação) Prof. Dr. Maycon Leone Maciel Peixoto (PGCOMP-UFBA)
Link para o curriculum lattes http://lattes.cnpq.br/5003713680310544
Data do exame 12 Dec, 2025
Horário do exame 1:00 PM

 

Data da Defesa: 
12/12/2025 - 13:00
Tipo de Defesa: 
Qualificação de Doutorado