Proactive autoscaling for high web traffic environments : a transformers-based approach | |
| Author | Coka, Somesh Rao |
| Call Number | AIT RSPR no.CS-23-05 |
| Subject(s) | Cloud computing Kubernetes Web applications--Development |
| Note | A research study submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science |
| Publisher | Asian Institute of Technology |
| Abstract | The evolution of software architecture patterns has been significantly influenced by the emergence of microservices and container-based technologies. These advancements have improved the modularity of applications and the development, testing, scaling, and component replacement processes. However, sophisticated orchestration capabilities are required to manage these applications, which can comprise hundreds of microservices with intricate interdependencies. Many application developers are currently drawn to the cloud computing environment and are deploying their web applications on cloud data centres. Cloud computing has become the preferred deployment environment for web applications. Kubernetes, widely used for deploying web applications on cloud platforms, offers an auto-scaling functionality to adapt to changing client needs. This research paper presents a novel proactive auto-scaling strategy for high-traffic web environments through a transformer-based model, which improves resource utilization efficiency and performance. By employing a month's worth of NASA server log data, our empirical analysis compares the Transformer model to the ARIMA and Random Forest models. This study provides valuable contributions to the field of predictive auto-scaling method selection strategy, showing the importance of customizing model selection to suit the specific requirements and constraints of particular applications. The findings are especially important for individuals working in the domains of auto-scaling and cloud computing since they pave the way for a more responsive and efficient cloud based service supply. |
| Year | 2023 |
| Type | Research Study Project Report (RSPR) |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Chantri Polprasert; |
| Examination Committee(s) | Chutiporn Anutariya;Chaklam Silpasuwanchai; |
| Scholarship Donor(s) | AIT Scholarships; |
| Degree | Research Studies Project Report (M. Eng.) - Asian Institute of Technology, 2023 |