Concurrency Control

Problem Definition
1. Objectives
  1. Throughput: number of transactions processed per unit time eg> 1 million transactions per second
  2. Maximum transactions throughput(work perform) while preventing interference among multiple users
2. Constraint
  1. No interference: result same as serial/sequential effect
  2. Interference occurs on commonly manipulated data known as hot spot.
Concurrency Control Problems
1. Last Update Problem
  1. Preventing Last Update Problem Using 2PL
2. Uncommitted Dependency Problem
  1. Preventing Uncommited using 2PL
Concurrency Control Mechanisms
1. Locking
  1. Obtain R/W Lock before accessing an item
  2. Locking Conflicts
  3. Locking Granularity
  4. Wait if a conflicting lock is held in The Lock Manager
    1. Shared Lock ( S lock): conflicts with exclusive locks
    2. Exclusive Lock (X lock): conflicts with all other kinds of locks
  5. Two-Phase Locking 2PL
    1. Growing Phase: All files are getting locked.
    2. Shrinking Phase: All locks are released
    3. Strict 2PL
    4. Deadlock
      1. Deadlock(Mutual Waiting)
      2. Deadlock Resolution
        Detection
        Overhead is reasonable for deadlocks among 2 or 3 transactions
        TimeOut
        Waiting Limit
        Can abort transactions that are not deadlock
        Timeout interval is difficult to determine
2. Optimistic Concurrency
  1. Private Workspaces Each process is executing privately

Resource summary

	Created by Muhammad Arshad about 1 year ago