<![CDATA[The telecommunications sector generates vast and rapidly growing data volumes, rendering traditional Relational Database Management Systems (RDBMS) insufficient for modern analytical demands. This paper investigates ATOMA's (Advanced Telecom Operations and Mobility of Afghanistan) strategic transition from a legacy Oracle-based data warehouse to a distributed Big Data ecosystem comprising Hadoop, Apache Spark, and Apache Kafka. Drawing on qualitative case study methodology, data were collected from 15 purposively selected IT professionals across eight functional teams using a structured questionnaire. Thematic analysis was conducted through the Technology-Organization-Environment (TOE) framework and the Migration Lifecycle Model. Findings reveal that ATOMA's primary migration drivers include Oracle scalability limitations, batch-reporting inefficiencies, missing Call Detail Records (CDRs), absence of real-time analytics, and cost reduction imperatives. Participants identified data migration complexity, skill gaps, system integration challenges, and change management as the most significant barriers. Anticipated benefits across all teams consistently highlighted real-time reporting, improved fraud detection, enhanced customer analytics, and open-source cost optimization. The paper proposes a Phased Big Data Adoption Framework (PBAF) tailored to telecom operators in fragile, resource-constrained environments comprising five stages: Assessment, Pilot, Hybrid Operation, Full-Scale Deployment, and Optimization. Findings are directly applicable to telecom operators in emerging markets facing analogous legacy-system migration challenges.]]>
Copyrights © 2026
