Lecture 12. Google Bigtable
Readings:
- Fay Chang, Jeffrey Dean, Sanjay Ghemawat, Wilson C. Hsieh, Deborah A. Wallach, Mike Burrows, Tushar Chandra, Andrew Fikes, and Robert E. Gruber. Bigtable: A Distributed Storage System for Structured Data. OSDI 2006. [pdf]
As you read this paper, consider the following questions:
- What is the goal of Bigtable?
- How is it similar and different from a DBMS (data model, API, features, implementation, etc.)?
- In Bigtable, data is stored in lexicographical order and clients can exploit this property when selecting their row keys to get good locality for their data accesses. Does this mean that Bigtable breaks physical data independence? Please discuss.
- Discuss the scalability properties of Bigtable.
Lecture notes:
Optional, additional readings: