Principles Of Distributed Database Systems Exercise Solutions

The Springer companion website for the textbook contains . However, these materials are restricted to instructors who have adopted the book for a course . If you are a course instructor, you can request access directly from Springer or through your university’s library.

The coordinator asks participants if they are ready to commit.

π_CustID(σ_City=‘Paris’(Customers)). Size: if 10% of customers in Paris → 500 CustIDs.

R = 10,000 tuples, S = 50,000 tuples. Hash function partitions data into 10 buckets. Each site sends its bucket to a single join site. Network cost = 1 per tuple. Local join cost negligible. Question: Compute total network cost. The Springer companion website for the textbook contains

is older. Under the protocol, an older transaction is allowed to wait for a younger one. T1cap T sub 1 T2cap T sub 2 requests a lock held by T1cap T sub 1 T2cap T sub 2 is younger ( ). The younger transaction dies. T2cap T sub 2 aborts, rolls back, releases its lock on , and clears the deadlock cycle. 4. Distributed Reliability and Commit Protocols

Total Cost=Cost1+Cost2=1,000+200,000=201,000 bytesTotal Cost equals Cost sub 1 plus Cost sub 2 equals 1 comma 000 plus 200 comma 000 equals 201 comma 000 bytes Conclusion

Often solved using optimization techniques, calculating the sum of query costs across sites. 3. Distributed Query Processing The coordinator asks participants if they are ready

Making distributed data appear as a single unit to the user.

You will trace the blocking vs. non-blocking behavior of the and Three-Phase Commit (3PC) protocols under different failure scenarios (e.g., site crashes, network partitions).

Always consider what happens when a site goes down midway through a protocol (e.g., during a 2PC prepare phase). Knowing how algorithms handle site failures (the blocking problem) is a favorite topic for exam questions. R = 10,000 tuples, S = 50,000 tuples

Distributed Database Systems (DDBS) aim to provide , reliability , and performance by spreading data across multiple sites. This write-up covers core principles often found in academic exercises based on the authoritative text by M. Tamer Özsu and Patrick Valduriez . 1. Data Distribution and Design

Basic TO rule for write_TS:

A staple exercise (e.g., ) asks you to give the algorithms for the transaction managers and lock managers for distributed two‑phase locking (2PL).

Explain how the Two-Phase Commit (2PC) protocol ensures data consistency. Solution: Phase 1: Voting Phase The Coordinator asks both Site X and Site Y to prepare.

Principles of Distributed Database Systems Exercise Solutions: A Comprehensive Guide