Evaluating Scalable Data Analytics Platforms in Cloud Environments For Real-Time Processing
B. Surekha,
Y. Sindhura,
A. Lakshmi Narayana
In the era of big data, cloud-based scalable data analytics platforms are critical for real-time processing
across various industries. This research provides a comparative evaluation of horizontal and vertical
scalability in cloud environments, focusing on their impact on performance, cost, and resource utilization.
Using platforms like AWS, Google Cloud, and Azure, we analyze key metrics such as CPU utilization,
latency, and operational costs across different scaling strategies. The results highlight the trade-offs
between horizontal scaling, which distributes workloads across multiple instances for improved latency
but at a higher cost, and vertical scaling, which optimizes resource usage within a single machine
but encounters performance limits as demand increases. Auto-scaling mechanisms are also examined
as a balance between these two approaches, offering dynamic scalability and cost optimization. This
study provides valuable insights into selecting the appropriate scaling strategy based on workload
requirements and operational constraints in real-time data analytics.
