CPS 216: Advanced Database Systems
(Data-Intensive Computing Systems, Fall 2009)

The initial readings for the in-class presentations on "Big Data" topics are posted below.

MapReduce Vs. Databases

• Article: MapReduce is a major step backwards (I)
• Article: MapReduce is a major step backwards (II)
• Article: Response to above articles
• Comparison of Parallel Databases Vs. MapReduce
• Article: Response to the above comparison

• Hive's approach
• Short Hive paper

Optimization of MapReduce: Plan Generation and Scheduling

• Automatic optimization
• Improving the scheduler

• Fair scheduling in a multi-user environment
• Further optimizations to scheduling

Databases over a Distributed FileSystem

• Google Bigtable
• Hbase

• Google FileSystem

Higher-level Languages for Parallel Data Processing

• Pig
• Scope

• Dryad LINQ paper
• Dryad LINQ project

Optimization of MapReduce: Data Placement and Rebalancing

• HDFS Design
• HDFS Rebalancer
• HadoopDB database system

• HDFS User Guide

Scalable Key-Value Stores

• Amazon's Dynamo system
• Yahoo!'s PNUTS system
• Article: Note on PNUTS

• Cassandra