Question 3: Assume that A, B, C, D, and E are data items, and that T1, T2, T3, T4, T5, and T6 are transactions. Assume that your DBMS is using shared/exclusive locks for concurrency control. Currently T1 has an exclusive lock on data item A, and T2 has a shared lock on data item C. All Use your knowledge in other data items are currently unlocked. concurrency control to answer the following questions. [around 6 minutes] Case (a) Consider the operations Read and Write. Which of these can T1 do it to A, given its lock (read, write, neither read nor write, or both read and write)? (b)Consider the operations Read and Write. Which of these can T2 do it to C, given its lock (read, write, neither read nor write, or both read and write)? (c) T3 requests an exclusive lock on D. Will T3 obtain the lock. at this point, or will T3 have to wait? (d) T4 requests a shared lock on C. Will T4 obtain the lock at this point, or will T4 have to wait? Answer

Question 3: Assume that A, B, C, D, and E are data items, and that T1, T2, T3, T4, T5, and T6 are transactions. Assume that your DBMS is using shared/exclusive locks for concurrency control. Currently T1 has an exclusive lock on data item A, and T2 has a shared lock on data item C. All Use your knowledge in other data items are currently unlocked. concurrency control to answer the following questions. [around 6 minutes] Case (a) Consider the operations Read and Write. Which of these can T1 do it to A, given its lock (read, write, neither read nor write, or both read and write)? (b)Consider the operations Read and Write. Which of these can T2 do it to C, given its lock (read, write, neither read nor write, or both read and write)? (c) T3 requests an exclusive lock on D. Will T3 obtain the lock. at this point, or will T3 have to wait? (d) T4 requests a shared lock on C. Will T4 obtain the lock at this point, or will T4 have to wait? Answer

Database System Concepts

7th Edition

ISBN:9780078022159

Author:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Publisher:Abraham Silberschatz Professor, Henry F. Korth, S. Sudarshan

Chapter1: Introduction

Section: Chapter Questions

Problem 1PE

See similar textbooks

Similar questions

Suppose the deferred modification technique is used in a database.a. Is the old value part of an update log record required any more?Why or why not? b. If old values are not stored in update log records, transaction undo isclearly not feasible. How would the redo phase of recovery have to bemodified as a result?c. Deferred modification can be implemented by keeping updated dataitems in local memory of transactions and reading data items that havenot been updated directly from the database buffer. Suggest how to efficientlyimplement a data item read, ensuring that a transaction sees itsown updates.d. What problem would arise with the above technique if transactions performa large number of updates?
Q. No. 2 Concurrency Control is a most important element for the proper functioning of a system where two or multiple database transactions that require access to the same data, are executed simultaneously. You have to discuss all issues of concurrent access of data with the help of self-assumed scenarios and data.
The following dependency graph shows the dependencies of four queries. Is a schedule from these queries conflict serializable? Yes 83 No 81 84 I don't have enough information to determine 82 Why is it important to have concurrency control in databases? Concurrency Control is needed so that multiple transactions can be executed at the same time. Concurrency Control is needed so that data can be updated correctly when multiple transactions are executed concurrently. Concurrency Control is needed to allow multiple users to edit the same record at the same time. O Concurrency Control is needed so that the database can schedule conflicts based on the priority of the transactions.
131. In concurrency control policy the the lock is obtained on a. Entire database b. A particular transaction alone c. All the new elements d. All of the mentioned
2. CHAPTER TRANSACTION MANAGEMENT Consider the case of immediate database approach and assume that, after a crash, the log contains the following records: Transaction Object Old Value New Value LSN L001 T1 (starts) P. 400 500 LO02 T1 300 400 LO03 T1 L004 T2 (starts) LO05 T1( commits) LO06 T3 (starts) R. 100 200 LO07 T2 700 800 LO08 T3 LO09 T4 (starts) L010 T3 (commits) T4 T. 800 900 L011 i) Which transaction(s) will be redone during recovery process? ii) Which transaction(s) will be undone during recovery process? ii) What will be the value of data item 'P' and 'T' after the recovery?
When an optimistic strategy for concurrency management is adopted, it may commonly cause transactions to wind up taking a very lengthy time. Explain?
When an optimistic method of concurrency management is used, it is not uncommon for transactions to take a significant amount of time to finish before being committed. Explain?
Concurrency control is a problem in Distributed Database Systems, just as it is in other distributed systems. Is it necessary to impose concurrency management techniques such as 2PL (Two-phase locking) in a situation where fragments are not replicated? Explain.
When an optimistic method of concurrency management is used, it is not unusual for transactions to take a considerable amount of time to finish. Explain.
7. Wei Han and Joon Liang are trying to register the lab session of Database Systems at different locations. The capacity of the lab session has only 1 remaining capacity. A transaction log file has captured this information: time, name of transaction, steps of command being performed, the stored value if any and whether or not the transaction is committed. Record of the log file is shown in the following table. Based on the log file, what is the problem of T1 and T2 at time 2 and 3? [single answer] * Committe Time Transaction Step Stored Value Read Capacity Write Capacity = 1-1 Read Capacity TI TI 1 No 3 T2 1 4 T! Rollback 5 T2 Write Capacity= 1-1 Yes Inconsistent retrieval Lost update Redundant data Data isolation Uncommitted data problem
Suppose the deferred modification technique is used in a database. a. Is the old-value part of an update log record required any more? Why or why not? b. If old values are not stored in update log records, transaction undo. Is clearly not feasible. How would the redo-phase of recovery have to be modified as a result? C. Deferred modification can be implemented by keeping updated data items in local memory of transactions, and reading data items that have not been updated directly from the database buffer. Suggest how to efficiently implement a data item read, ensuring that a trans action sees its own updates.
Many applications need to generate sequence numbers for each transaction. a. If a sequence counter is locked in two-phase manner, it can become a concurrency bottleneck. Explain why this may be the case. b. Many database systems support built-in sequence counters that are not locked in two-phase manner; when a transaction requests a sequence number, the counter is locked, incremented and unlocked. i. Explain how such counters can improve concurrency. ii. Explain why there may be gaps in the sequence numbers belonging to the final set of committed transactions.