DatabaseSystemRecovery.pptVIP

Database System Recovery CSE544 Philip A. Bernstein Copyright ?2000 Philip A. Bernstein Outline 1. Introduction 2. DB Recovery Model 3. Recovery Manager 4. Log-based Recovery 5. Media Failure 6. Shadow Paging 1. Introduction Transaction - The execution of a program that performs an administrative function by accessing a shared database, usually on behalf of an on-line user. Transaction Processing – The technology for building large-scale systems that run transactions Examples What Makes Transaction Processing (TP) Hard? Atomicity - no partial results Isolation – appears to run transactions serially Durability – results aren’t lost by a failure Availability - system must be up all the time Reliability - system should rarely fail Response time - within 1-2 seconds Throughput - thousands of transactions/second Scalability - start small, ramp up to Internet-scale Configurability - for above requirements + low cost Distribution - of users and data Example: An E-Commerce Server Zooming In At least half of the work of making a TP system work is outside the transaction boundary I.e., above the yellow line on the previous slide Much of it is distributed computing Some is TP-specific (e.g. persistent queues, workflow) This section of the course focuses below the yellow line, in the DBMS Focusing on Recovery All database reads/writes are within a transaction Transactions have the “ACID” properties Atomicity - all or nothing Consistency - preserves database integrity Isolation - execute as if they were run alone Durability - results aren’t lost by a failure Recovery subsystem guarantees A D Concurrency control guarantees I Application program guarantees C 2. DB Recovery Model A database may become inconsistent because of a transaction failure (abort) database system failure (possibly caused by OS crash) media crash (disk-resident data is corrupted) The recovery system ensures the database contains exactly those updates produced by committed transactions I.e. atomicity an