Big Data Analytics and Applications

Daffodil International University Department of Computing and Information Systems Course: Big Data Analysis and Applications — Course Code: CIS 312 Class Test 01 — Time: 30 Mins — Total Marks: 15 — SET – A 1. (a) Differentiate between Scale-Up (Vertical Scaling) and Scale-Out (Horizontal Scaling) architectures in the context of Big Data processing. [3] (b) Why does Scale-Up reach a physical and economic ceiling when transitioning from Gigabyte-scale to Petabyte-scale workloads? [2] 2. (a) Compare Schema-on-Write and Schema-on-Read strategies across the parameters of ingestion speed, structural flexibility, and read performance. [3] (b) Describe a real-world scenario where applying Schema-on-Read provides a clear architectural advantage over traditional Schema-on-Write. [2] 3. (a) Explain the core principle of "Data Locality" in distributed computing. Why is moving computation code to data nodes vastly more efficient than moving data to computation nodes? [3] (b) Identify two primary reasons why strict ACID transactional guarantees in traditional relational databases become performance bottlenecks at Big Data scale. [2]