Cloud-Native Student Information Management: A Serverless NoSQL Architecture with Infrastructure-as-Code and Continuous Deployment on AWS

Abstract: Higher education institutions increasingly face challenges in managing fragmented student data across disparate systems, resulting in operational inefficiencies and delayed access to academic information. This paper presents the design, implementation, and evaluation of a cloud-native Student Information System (SIS) built on a fully serverless NoSQL backend using Amazon Web Services (AWS) DynamoDB, provisioned through AWS CloudFormation Infrastructure-as-Code (IaC) templates, and delivered via an automated continuous integration and continuous deployment (CI/CD) pipeline. The system employs Python and Flask as the application framework, integrates Flask-SocketIO for real-time WebSocket-based data broadcasting, and enforces security through bcrypt- based password hashing and session management via Flask-Login. A five-table DynamoDB schema operating in PAY_PER_REQUEST billing mode eliminates capacity planning overhead while ensuring high availability. The infrastructure stack provisions a Virtual Private Cloud (VPC), EC2 compute instance, IAM roles, security groups, and an Elastic IP through a single declarative CloudFormation template. An AWS CodePipeline orchestrates CodeBuild and CodeDeploy stages triggered automatically by CodeCommit repository events via Amazon EventBridge. Experimental evaluation demonstrates sub-100 ms response times for core academic operations and WebSocket event delivery latencies below 30 ms. The proposed architecture reduces manual deployment effort and infrastructure drift, providing an open, replicable template for cloud-native academic information systems in resource-constrained institutional contexts.

Keywords: Student Information System; AWS DynamoDB; CloudFormation; Infrastructure-as-Code; CI/CD; Flask; WebSocket; SocketIO; Cloud-Native; Serverless NoSQL.