Cloud Case Studies: Apache Hadoop and Aneka Cloud Application Platform
In university exams, you will often find dedicated questions on two open-source and enterprise distributed computing frameworks: Apache Hadoop and Aneka.
1. Case Study 1: Apache Hadoop (Big Data Cloud Engine)
Apache Hadoop is an open-source software framework designed to store and process gigantic volumes of data (terabytes and petabytes) across clusters of cheap commodity computers.
- Splits huge files into fixed 128MB chunks.
- Distributes blocks across multiple servers.
- 3x Replication: Keeps 3 copies of every block automatically across different server racks.
- Architecture: Single NameNode (Master) + Multiple DataNodes (Slaves).
- Map Phase: Breaks a large computation into hundreds of small tasks executed where the data lives.
- Reduce Phase: Aggregates, sorts, and combines results from all workers into a final answer.
- Data Locality: Moves code to the data instead of moving massive data over slow network cables.
2. Case Study 2: Aneka Cloud Application Platform
Developed by the Cloud Computing and Distributed Systems (CLOUDS) Laboratory at the University of Melbourne under Professor Rajkumar Buyya, Aneka is a pure enterprise cloud software platform for building and deploying distributed applications.
Architecture of Aneka:
Aneka operates as a service-oriented platform built on Microsoft .NET. It consists of:
- Aneka Container: A lightweight software runtime installed on every physical or virtual server in the cluster.
- Fabric Services: Responsible for discovering machines, monitoring hardware, and managing dynamic resource allocation.
- Execution Services: Schedules and runs tasks according to three supported programming models:
- Task Model: Independent bag-of-tasks.
- Thread Model: Multi-threaded distributed execution.
- MapReduce Model: Large-scale distributed data processing.
- Accounting & Billing Services: Measures CPU seconds and charges users based on market economic models.
3. Comparison: Hadoop vs. Aneka
| Comparison Point | Apache Hadoop | Aneka Platform |
|---|---|---|
| Primary Purpose | Batch Big Data analytics and distributed storage | Multi-model distributed application development |
| Underlying Tech | Java / Apache Open Source | Microsoft .NET Framework / C# |
| Supported Models | MapReduce, Hive, Spark | Task, Thread, and MapReduce models |
| Storage Engine | HDFS (Hadoop Distributed File System) | Relies on external databases or cloud storage |
| Commercial Use | Big data mining in Yahoo, Facebook, Netflix | Enterprise grid & private cloud research deployments |
💡 DAVV Exam Tip
Question (10 Marks): “Explain the architecture of Apache Hadoop and Aneka Cloud Platform with neat diagrams.”
- For Hadoop: Draw the NameNode / DataNodes architecture and explain 3x replication and MapReduce.
- For Aneka: Mention Prof. Rajkumar Buyya, .NET container, and the 3 programming models (Task, Thread, MapReduce). `;
fs.writeFileSync(path.join(baseDir, ‘case-studies-hadoop-and-aneka.md’), topic5, ‘utf8’);