Chapter 17 recovery system
Download
1 / 121

Chapter 17: Recovery System - PowerPoint PPT Presentation


  • 133 Views
  • Uploaded on

Chapter 17: Recovery System. Database System Concepts. Chapter 1: Introduction Part 1: Relational databases Chapter 2: Relational Model Chapter 3: SQL Chapter 4: Advanced SQL Chapter 5: Other Relational Languages Part 2: Database Design

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' Chapter 17: Recovery System' - gerard


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

Database system concepts
Database System Concepts

  • Chapter 1: Introduction

  • Part 1: Relational databases

    • Chapter 2: Relational Model

    • Chapter 3: SQL

    • Chapter 4: Advanced SQL

    • Chapter 5: Other Relational Languages

  • Part 2: Database Design

    • Chapter 6: Database Design and the E-R Model

    • Chapter 7: Relational Database Design

    • Chapter 8: Application Design and Development

  • Part 3: Object-based databases and XML

    • Chapter 9: Object-Based Databases

    • Chapter 10: XML

  • Part 4: Data storage and querying

    • Chapter 11: Storage and File Structure

    • Chapter 12: Indexing and Hashing

    • Chapter 13: Query Processing

    • Chapter 14: Query Optimization

  • Part 5: Transaction management

    • Chapter 15: Transactions

    • Chapter 16: Concurrency control

    • Chapter 17: Recovery System

  • Part 6: Data Mining and Information Retrieval

    • Chapter 18: Data Analysis and Mining

    • Chapter 19: Information Retreival

  • Part 7: Database system architecture

    • Chapter 20: Database-System Architecture

    • Chapter 21: Parallel Databases

    • Chapter 22: Distributed Databases

  • Part 8: Other topics

    • Chapter 23: Advanced Application Development

    • Chapter 24: Advanced Data Types and New Applications

    • Chapter 25: Advanced Transaction Processing

  • Part 9: Case studies

    • Chapter 26: PostgreSQL

    • Chapter 27: Oracle

    • Chapter 28: IBM DB2

    • Chapter 29: Microsoft SQL Server

  • Online Appendices

    • Appendix A: Network Model

    • Appendix B: Hierarchical Model

    • Appendix C: Advanced RelationalDatabase Model


Part 5 transaction management chapters 15 through 17
Part 5: Transaction management (Chapters 15 through 17).

  • Chapter 15: Transactions

    • focuses on the fundamentals of a transaction-processing system, including transaction atomicity, consistency, isolation, and durability, as well as the notion of serializability.

  • Chapter 16: Concurrency control

    • focuses on concurrency control and presents several techniques for ensuring serializability, including locking, timestamping, and optimistic (validation) techniques. The chapter also covers deadlock issues.

  • Chapter 17: Recovery System

    • covers the primary techniques for ensuring correct transaction execution despite system crashes and disk failures. These techniques include logs, checkpoints, and database dumps.


Chapter 17 recovery system1

17.1 Failure Classification

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Failure classification

Transaction failure:

Logical errors: transaction cannot complete due to some internal error condition

System errors: the database system must terminate an active transaction due to an error condition (e.g., deadlock)

System crash: a power failure or other hardware or software failure causes the system to crash (halt).

Fail-stop assumption: non-volatile storage contents are assumed to not be corrupted by system crash

Database systems have numerous integrity checks to prevent corruption of disk data

Disk failure: a head crash or similar disk failure destroys all or part of disk storage

Destruction is assumed to be detectable: disk drives use checksums to detect failures

Multiple copies or archival tapes are solutions

Failure Classification


Recovery algorithms
Recovery Algorithms

  • Recovery algorithms are techniques to ensure database consistency and transaction atomicity and durability despite failures

    • Focus of this chapter

  • Recovery algorithms have two parts

    • Actions taken during normal transaction processing to ensure enough information exists to recover from failures

    • Actions taken after a failure to recover the database contents to a state that ensures atomicity, consistency and durability


Chapter 17 recovery system2

17.1 Failure Classification

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Storage structure

Volatile storage:

does not survive system crashes

examples: main memory, cache memory

Nonvolatile storage:

survives system crashes

examples: disk, tape, flash memory, non-volatile (battery backed up) RAM

Stable storage:

a mythical form of storage that survives all failures

approximated by maintaining multiple copies on distinct nonvolatile media

for log space

Storage Structure


Stable storage implementation

Maintain multiple copies of each block on separate disks

copies can be at remote sites to protect against disasters such as fire or flooding.

Failure during data transfer can still result in inconsistent copies.

Block transfer can result in

Successful completion

Partial failure: destination block has incorrect information

Total failure: destination block was never updated

Protecting storage media from failure during data transfer (one solution):

Execute output operation as follows (assuming two copies of each block):

Write the information onto the first physical block.

When the first write successfully completes, write the same information onto the second physical block.

The output is completed only after the second write successfully completes.

Stable-Storage Implementation


Stable storage implementation cont

Protecting storage media from failure during data transfer (cont.):

If copies of a block differ due to failure during output operation  To recover from failure:

First find inconsistent blocks:

Expensive solution: Compare the two copies of every disk block.

Better solution:

Record in-progress disk writes on non-volatile storage (Non-volatile RAM or special area of disk).

Use this information during recovery to find blocks that may be inconsistent, and only compare copies of these.

Used in hardware RAID systems

If either copy of an inconsistent block is detected to have an error (bad checksum), overwrite it by the other copy. If both have no error, but are different, overwrite the second block by the first block.

Stable-Storage Implementation (Cont.)


Data access

Physical blocks (cont.): are those blocks residing on the disk.

Buffer blocks are the blocks residing temporarily in main memory.

Block movements between disk and main memory are initiated through the following two operations:

input(B) transfers the physical block B to main memory.

output(B) transfers the buffer block B to the disk, and replaces the appropriate physical block there.

Each transaction Tihas its private work-area in which local copies of all data items accessed and updated by it are kept.

Ti's local copy of a data item X is called xi.

We assume, for simplicity, that each data item fits in, and is stored inside, a single block.

Data Access



Data access cont

Transaction transfers data items between (cont.):system buffer blocks and its private work-area using the following operations :

read(X) assigns the value of data item X to the local variable xi.

write(X) assigns the value of local variable xito data item {X} in the buffer block.

both these commands may necessitate the issue of an input(BX) instruction before the assignment, if the block BX in which X resides is not already in memory.

Transactions

Perform read(X) while accessing X for the first time;

All subsequent accesses are to the local copy.

After last access, transaction executes write(X).

output(BX) need not immediately follow write(X).

System can perform the output operation when it deems fit.

Data Access (Cont.)


Example of data access
Example of Data Access (cont.):

buffer

input(A)

Buffer Block A

x

A

Buffer Block B

Y

output(B)

B

read(X)

write(Y)

disk

x2

x1

y1

work area

of T2

work area

of T1

memory


Chapter 17 recovery system3

17.1 Failure Classification (cont.):

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Recovery and atomicity

Modifying the database without ensuring that the transaction will commit may leave the database in an inconsistent state

Consider transaction Ti that transfers $50 from account A to account B

Goal is either to perform all database modifications made by Tior none at all.

Several output operations may be required for Ti (to output A and B).

A failure may occur after one of these modifications have been made but before all of them are made.

To ensure atomicity despite failures, we first output information describing the modifications to stable storage without modifying the database itself.

We study two approaches:

log-based recovery

shadow-paging

We assume (initially) that transactions run serially, that is, one after the other.

Recovery and Atomicity


Chapter 17 recovery system4

17.1 Failure Classification will commit may leave the database in

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Log based recovery

A will commit may leave the database in log is kept on stable storage.

The log is a sequence of log records, and maintains a record of update activities on the database.

Log Records

When transaction Tistarts, it registers itself by writing a <Tistart> log record

Before Tiexecutes write(X), a log record <Ti, X, V1, V2>is written, where V1 is the value of X before the write, andV2is the value to be written to X.

Log record notes that Ti has performed a write on data item X.

X had value V1before the write, and will have value V2after the write.

When Tifinishes it last statement, the log record <Ticommit> is written.

We assume for now that log records are written directly to stable storage (that is, they are not buffered)

Two approaches using logs

Deferred database modification

Immediate database modification

Log-Based Recovery

At the moment, Assume serial execution of Transactions T0, T1, T3…


Example of data access1
Example of Data Access will commit may leave the database in

buffer

input(A)

Buffer Block A

x

Database

A

Buffer Block B

Y

output(B)

B

read(X)

write(Y)

Disk

x2

x1

Log file

y1

Log records

work area

of T2

work area

of T1

Main memory


Deferred database modification

The will commit may leave the database in deferred database modification scheme records all modifications to the log, but defers all the writes to after partial commit.

Assume that transactions execute serially

Transaction starts by writing <Tistart>record to log.

A write(X) operation results in a log record <Ti, X, V>being written

where V is the new value for X

Note: old value is not needed for this scheme

Deferred Update Steps

“write” operationis not performed on X at this time, but is deferred.

When Tipartially commits, <Ticommit> is written to the log

Finally, the log records are read and used to actually execute the previously deferred writes.

Deferred Database Modification


Deferred database modification cont

REDO only scheme will commit may leave the database in

During recovery after a crash, a transaction needs to be redone if and only if both <Tistart> and <Ti commit> are there in the log.

Redoing a transaction Ti( redoTi) sets the value of all data items updated by the transaction to the new values.

Crashes can occur while

the transaction is executing the original updates, or

while recovery action is being taken

Deferred Database Modification (Cont.)


State of the log and database corresponding to t 0 and t 1

Deferred Database Modification: State of Log and Database will commit may leave the database in

State of the Log and Database corresponding to T0 and T1

  • Example transactions T0and T1(T0executes before T1):

    T0: read (A) T1: read (C)

    A = A - 50C = C - 100

    Write (A) write (C)

    read (B)

    B = B + 50

    write (B)


Deferred database modification cont1

Below we show the log as it appears at three instances of time.

If log on stable storage at time of crash is as in case:

(a) No redo actions need to be taken

(b) redo(T0) must be performed since <T0 commit> is present

(c) { redo(T0) ; redo(T1)} since <T0commit> and < T1 commit> are present

Redo 하는 이유: commit이 되어도 buffer까지 update이 된것은 알지만 DB에 update가 되었는지 안되었는지 지안된지는 알수 없으므로, buffer에 update값을 살려놔야 안심할수 있다.

Deferred Database Modification (Cont.)

T0, T1, T2 …..


Immediate database modification

The time.immediate database modification scheme allows database updates of an uncommitted transaction to be made as the writes are issued

Since undoing may be needed, update logs must have both old value and new value

Update log record must be written before database item is written

We assume that the log record is output directly to stable storage

Before execution of an output(B) operation for a data block B, all log records corresponding to items B must be flushed to stable storage

Output of updated blocks can take place at any time before or after transaction commit

Order in which blocks are output can be different from the order in which they are written in the buffer.

Immediate Database Modification


Immediate database modification example

Log Write Output time.

<T0start>

<T0, A, 1000, 950>

<To, B, 2000, 2050>

A = 950

B = 2050

<T0commit>

<T1start>

<T1, C, 700, 600>

C = 600

BB, BC

<T1commit>

BA

Note: BXdenotes block containing X.

Immediate Database Modification Example

  • Example transactions T0and T1

    (T0executes before T1):

    T0: read (A)

    C = C - 100

    Write (A)

    read (B)

    B = B + 50

    write (B)

    T1: read (C)

    A = A – 50

    write (C)


Immediate database modification cont

Recovery procedure has two operations instead of one: time.

undo(Ti) restores the value of all data items updated by Tito their old values, going backwards from the last log record for Ti

redo(Ti) sets the value of all data items updated by Tito the new values, going forward from the first log record for Ti

Both operations must be idempotent

undo(Ti) = undo ( undo(Ti) ) = undo (undo (undo(Ti) ))…….

That is, even if the operation is executed multiple times the effect is the same as if it is executed once

Needed since operations may get re-executed during recovery

When recovering after failure:

Transaction Tineeds to be undone if the log contains the record <Tistart>, but does not contain the record <Ticommit>.

Transaction Tineeds to be redone if the log contains both the record <Tistart> and the record <Ti commit>.

Undo operations are performed first, then redo operations.

Immediate Database Modification (Cont.)


Immediate db modification recovery example

Below we show the log as it appears at three instances of time.

Recovery actions in each case above are:

(a) undo (T0): B is restored to 2000 and A to 1000.

(b) undo (T1) and redo (T0): C is restored to 700, and then A and B are

set to 950 and 2050 respectively.

redo (T0) and redo (T1): A and B are set to 950 and 2050 respectively. Then C is set to 600

* Undo하는 이유: buffer에 반영이 되었었으므로 혹시 DB에도 반영되었을 가능성도 있으므로, 다시 buffer에 예전값을 써주어서 DB에 반영이 되었어도 다시 원상복귀시키기위해.

Immediate DB Modification Recovery Example

Assume T0, T1, T2….


Checkpoints

Problems in recovery procedure as discussed earlier : time.

searching the entire log is time-consuming

we might unnecessarily redo transactions which have already output their updates to the database.

Streamline recovery procedure by periodically performing checkpointing

Output all log records currently residing in main memory onto stable storage.

Output all modified buffer blocks to the disk.

Write a log record < checkpoint> onto stable storage.

During the checkpointing, other transactions cannot be processed

Checkpoints

checkpoint

Log records

buffer

log

M1

A

Insert <checkpoint> into log


Checkpoints cont

During recovery we need to consider time.only the most recent transaction Ti that started before the checkpoint, and transactions that started after Ti.

Scan backwards from end of log to find the most recent <checkpoint> record

Continue scanning backwards till a record <Ti start> is found.

Need only consider the part of log following above start record.

Earlier part of log can be ignored during recovery, and can be erased whenever desired.

For all transactions (starting from Ti or later) with no <Ticommit>, execute undo(Ti). (Done only in case of immediate modification.)

Scanning forward in the log, for all transactions starting from Tior later with a <Ticommit>, execute redo(Ti).

Checkpoints (Cont.)


Example of checkpoints

T time.1 can be ignored (updates already output to disk due to checkpoint)

Undo T4 (remember Undo first, then Redo) // 이유는 clear하지 않다

Redo T2 and T3.

Example of Checkpoints

T1, T2, T3, T4….

Tf

Tc

T1

T2

T3

T4

system failure

checkpoint

  • Be careful for the order of undo & redo

  • Suppose T1 (x =3 ); T2 (x = x + 1); T3 (x= x +1); T4 (x = x * 2)


Chapter 17 recovery system5

17.1 Failure Classification time.

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery With Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Shadow paging

Shadow paging time. is an alternative to log-based recovery

this scheme is useful if transactions execute serially

Idea: maintain two page tables during the lifetime of a transaction

the current page table, and the shadow page table

Store the shadow page table in nonvolatile storage, such that state of the database prior to transaction execution may be recovered.

Shadow page table is never modified during execution

To start with, both the page tables are identical.

Only current page table is used for data item accesses during execution of the transaction.

Whenever any page is about to be written for the first time

A copy of this page is made onto an unused page.

The current page table is then made to point to the copy

The update is performed on the copy

Shadow Paging


Example of shadow paging
Example of Shadow Paging time.

Shadow and current page tables after write to page 4


Shadow paging cont

To commit a transaction : time.

1. Flush all modified pages in main memory to disk

2. Output current page table to disk

3. Make the current page table the new shadow page table, as follows:

keep a pointer to the shadow page table at a fixed (known) location on disk.

Simply update the pointer to point to current page table on disk

Once pointer to shadow page table has been written, transaction is committed.

No recovery is needed after a crash — new transactions can start right away, using the shadow page table.

Pages not pointed to from current/shadow page table should be freed (garbage collected).

Shadow Paging (Cont.)


Shadow paging cont1

Advantages of shadow-paging over log-based schemes time.

no overhead of writing log records

recovery is trivial

Disadvantages :

Copying the entire page table is very expensive

Can be reduced by using a page table structured like a B+-tree

No need to copy entire tree, only need to copy paths in the tree that lead to updated leaf nodes

Commit overhead is high even with above extension

Need to flush every updated page, and page table

Data gets fragmented (related pages get separated on disk)

After every transaction completion, the database pages containing old versions of modified data need to be garbage collected

Hard to extend algorithm to allow transactions to run concurrently

Easier to extend log based schemes

Shadow Paging (Cont.)


Chapter 17 recovery system6

17.1 Failure Classification time.

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Recovery with concurrent transactions

We modify the log-based recovery schemes time.to allow multiple transactions to execute concurrently.

All transactions share a single disk buffer and a single log

A buffer block can have data items updated by one or more transactions

We assume concurrency control using strict two-phase locking;

i.e. the updates of uncommitted transactions should not be visible to other transactions

Otherwise how to perform undo if T1 updates A, then T2 updates A and commits, and finally T1 has to abort?

Logging is done as described earlier.

Log records of different transactions may be interspersed in the log.

The checkpointing technique and actions taken on recoveryhave to be changed

since several transactions may be active when a checkpoint is performed.

Intersperse: 산재해 있다

Recovery with Concurrent Transactions


Recovery with concurrent transactions cont

Checkpoints are performed as before, except that the checkpoint log record is now of the form < checkpointL>where L is the list of transactions active at the time of the checkpoint

We assume no updates are in progress while the checkpoint is carried out (will relax this later)

When the system recovers from a crash, it first does the following:

1. Initialize undo-list and redo-list to empty

2. Scan the log backwards from the end, stopping when the first <checkpointL> record is found. For each record found during the backward scan:

if the record is <Ticommit>, add Titoredo-list

if the record is <Ti start>, then if Ti is not in redo-list, add Ti toundo-list

3. For every Ti in L, if Ti is not in redo-list, add Titoundo-list

Recovery with Concurrent Transactions (Cont.)


Recovery with concurrent transactions cont1

At this point checkpoint log record is now of the form undo-list consists of incomplete transactions which must be undone, and redo-listconsists of finished transactions that must be redone.

4. Recovery now continues as follows:

Scan log backwards from most recent record, stopping when <Tistart> records have been encountered for every Ti in undo-list.

During the scan, perform undo for each log record that belongs to a transaction in undo-list.

Locate the most recent <checkpoint L> record.

Scan log forwards from the <checkpoint L> record till the end of the log.

During the scan, perform redo for each log record that belongs to a transaction on redo-list

Recovery with Concurrent Transactions (Cont.)

Undo first, Redo next


Example of recovery with concurrent transactions

Go over the steps of the recovery algorithm on the following log:

<T0start>

<T0, A, 0, 10>

<T0commit>

<T1start>

<T1, B, 0, 10>

<T2start>

<T2, C, 0, 10>

<T2, C, 10, 20>

<checkpoint {T1, T2}>

<T3start>

<T3, A, 10, 20>

<T3, D, 0, 10>

<T3commit> /* the end of log */

Example of Recovery with concurrent transactions

  • In the step 1, 2, and 3

    • Undo-list = { T2 , T1}

    • Redo-list = { T3 }

  • In the step 4

    • Undo T2 , Undo T1 ,then Redo T3

T1

T2

T3

T0

crash

checkpoint

Crash here


Chapter 17 recovery system7

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Log record buffering

Log record buffering following log:: log records are buffered in main memory, instead of being output directly to stable storage.

Log records are output to stable storage when a block of log records in the buffer is full, or a log force operation is executed.

Log force is performed to commit a transaction by forcing all its log records (including the commit record) to stable storage.

Several log records can thus be output using a single output operation

reducing the I/O cost.

The rules below must be followed if log records are buffered:

Write-ahead logging or WALrule

Before a block of data in main memory is output to the database, all log records pertaining to data in that block must have been output to stable storage.

Log records are output to stable storage in the order in which they are created.

Transaction Ti enters the commit state only when the log record <Ticommit> has been output to stable storage.

Log Record Buffering


Database buffering

Database maintains following log:an in-memory buffer of data blocks

When a new block is needed and the buffer is full, an existing block needs to be removed from buffer

If the block chosen for removal has been updated, it must be output to disk

As a result of the WAL rule, if a block with uncommitted updates is output to disk, log records with undo information for the updates are output to the log on stable storagefirst.

<T0start>

<T0, A, 1000, 950>

Transaction T0 issues read(B)

Suppose the block (having A) is chosen to be output to disk for the block (having B) to come to the memory.  <T0, A, 1000, 950> should be output to a stable storage first

No updates should be in progress on a block when it is output to disk.

Before writing a data item, transaction acquires exclusive lock on block containing the data item

Lock can be released once the write is completed.

Such locks held for short duration are called latches.

Database Buffering


Os role in database buffering

Database buffer can be implemented either following log:

in an area of real main-memory reserved for the database

in virtual memory by OS

Implementing buffer in reserved main-memory has drawbacks:

Memory is partitioned before-hand between database buffer and applications, limiting flexibility.

DB applications and Non-DB applications cannot share the unused main-memory area

Needs may change, and although operating system knows best how memory should be divided up at any time, it cannot change the partitioning of memory.

OS Role in Database Buffering


Os role in database buffering cont

Database buffers are generally implemented in following log:virtual memory in spite of some drawbacks:

When OS needs to evict a page that has been modified, to make space for another page, the page is written to swap space on disk.

When database decides to write buffer page to disk, buffer page may be in swap space, and may have to be read from swap space on disk and output to the database on disk, resulting in extra I/O!

Known as dual paging problem.

Ideally when swapping out a database buffer page, OS should pass control to database, which in turn outputs page to database instead of to swap space (making sure to output log records first)

Dual paging can thus be avoided, but common OSs do not support such functionality.

OS Role in Database Buffering (Cont.)


Chapter 17 recovery system8

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Failure with loss of nonvolatile storage

So far we assumed no loss of non-volatile storage following log:

Technique similar to checkpointing used to deal with loss of non-volatile storage

Periodically dump the entire content of the database to stable storage

No transaction may be active during the dump procedure; a procedure similar to checkpointing must take place

Output all log records currently residing in main memory onto stable storage.

Output all buffer blocks onto the disk.

Copy the contents of the database to stable storage.

Output a record <dump> to log on stable storage.

To recover from disk failure

restore database from the most recent dump.

Consult the log and redo all transactions that committed after the dump

Can be extended to allow transactions to be active during dump; known as fuzzy dump or online dump

Will study fuzzy checkpointing later

Failure with Loss of Nonvolatile Storage


예제추가 following log:


Chapter 17 recovery system9

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery with Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Advanced recovery techniques

The recovery scheme in section 17.5 assumes strict 2PL following log:

Strict 2PL is not suitable for B+ tree index pages

Practically and widely B+ tree concurrency control

Early lock releases in B+ tree concurrency control algorithm does not honor 2PL

Therefore the recovery scheme in section 17.5 cannot be used in the real situations

In 1990, C. Mohan in IBM published the ARIES technique

Later most commercial DBMS adopted either ARIES or ARIES variants

Advanced Recovery Techniques is the summary of ARIES or ARIES variants

Advanced Recovery Techniques


Advanced recovery techniques logical undo logging 1

Support high-concurrency locking techniques following log:, such as those used for B+-tree concurrency control

Operations like B+-tree insertions and deletions release locks early.

They cannot be undone by restoring old values (physical undo), since once a lock is released, other transactions may have updated the B+-tree.

Instead, insertions (resp. deletions) are undone by executing a deletion (resp. insertion) operation (known as logical undo).

For such operations, undo log records should contain the undo operation to be executed

called logical undo logging, in contrast to physical undo logging.

Redo information is logged physically (that is, new value for each write) even for such operations

Logical redo is very complicated since database state on disk may not be “operation consistent”

Instead, physically redoing all updates of all transactions after the last checkpoint

Advanced Recovery Techniques:Logical Undo Logging (1)


Advanced recovery techniques logical undo logging 2

Operation logging following log: is done as follows:

When operation starts, log <Ti, Oj, operation-begin>. Here Oj is a unique identifier of the operation instance.

While operation is executing, normal log records with physical redo and physical undo information are logged.

When operation completes, <Ti, Oj, operation-end, U > is logged, where Ucontains information needed to perform a logical undo information.

If crash/rollback occurs before operation completes:

the operation-end log record is not found, and

the physical undo information is used to undo operation.

If crash/rollback occurs after the operation completes:

the operation-end log record is found, and in this case

logical undo is performed using U; the physical undo information for the operation is ignored.

Redo of operation (after crash) still uses physical redo information by physically redoing all updates of all transactions,

Advanced Recovery Techniques:Logical Undo Logging (2)


Advanced recovery techniques transaction rollback 1

Rollback of transaction following log:Tiis done as follows:

Scan the log backwards

If a log record <Ti, X, V1, V2> is found, perform the undo and log a special redo-only log record <Ti, X, V1>.

If a <Ti, Oj, operation-end, U> record is found

Rollback the operation logically using the undo information U.

Updates performed during roll back are logged just like during normal operation execution.

At the end of the operation rollback, instead of logging an operation-end record, generate a record

<Ti, Oj, operation-abort>.

Skip all preceding log records for Ti until the record <Ti, Ojoperation-begin> is found

Advanced Recovery Techniques:Transaction Rollback (1)


Advanced recovery techniques transaction rollback 2

Scan the log backwards (cont.): following log:

If a redo-only record is found ignore it

If a <Ti, Oj,operation-abort > record is found:

skip all preceding log records for Ti until the record <Ti, Oj, operation-begin > is found.

Stop the scan when the record <Ti, start> is found

Add a <Ti, abort> record to the log

Some points to note:

Cases 3 and 4 above can occur only if the database crashes while a transaction is being rolled back.

Skipping of log records as in case 4 is important to prevent multiple rollback of the same operation.

Advanced Recovery Techniques:Transaction Rollback (2)


Transaction rollback example under the early lock release

Log following log:

<T0start>

<T0, OA, operation-begin>

<T0, A, 950, 1000>

<T0, OA, operation-end, A = A + 50>

<T0, OB, operation-begin>

<To, B, 2050, 2000>

<To, B, 2000>

<T0, OA’, operation-begin>

<T0, A, 1000, 950>

<T0, OA’, operation-end, A = A - 50>

<T0, OA, operation-abort>

< T0abort>

Transaction Rollback Example(under the early lock release)

  • Example transactions T0

    T0: read (A)

    A = A - 50

    write (A)

    read (B)

    B = B + 50

    write (B)

rollback

Rollback

occurs

Log records are generated by scan the log backward after rollback

A = A – 50 에 대한 Undo는 A에 50을 더해주는 것이 되어야지 A = A – 50 을 수행하기 전 A값으로 돌아가는 것으로 충분하지 않다. 다른 transaction이 A값을 변경해 놓을 수도 있다.


Advanced recovery techniques restart recovery 1

The following actions are taken following log: when recovering from system crash

Scan log forward from last < checkpoint L> record

Repeat history by physically redoing all updates of all transactions,

Create an undo-list during the scan as follows

undo-list is set to L initially

Whenever <Tistart> is found Tiis added to undo-list

Whenever <Ticommit> or <Tiabort> is found, Ti is deleted from undo-list

Step1.1 brings database to state as of crash, with committed as well as uncommitted transactions having been redone. (redo-list is not necessary)

In step 1.2 undo-list contains transactions that areincomplete, that is, have neither committed nor been fully rolled back.

Advanced Recovery Techniques:Restart Recovery (1)


Advanced recovery techniques restart recovery 2

Recovery from system crash (cont.) following log:

Scan log backwards, performing undo on log records of transactions found in undo-list.

Transactions are rolled back as described earlier.

When <Ti start> is found for a transaction Ti in undo-list, write a <Tiabort> log record.

Stop scan when <Tistart> records have been found for all Ti in undo-list

Step 2 undoes the effects of incomplete transactions (those with neither commit nor abort log records). Recovery is now complete.

Advanced Recovery Techniques:Restart Recovery (2)


Recovery from system crash example
Recovery from System Crash Example following log:

Log Undo-list

<checkpoint L> { }

<T1start> {T1}

<T1commit> { }

<T2start> {T2}

<T3start> {T2, T3}

<T2abort> {T3}

Step 1: Scan log forward;

Redo is performed naturally

System

crash!

Step 2: Scan log backward;

Only T3 is rolled back as described earlier (as in 17.55)


Advanced recovery techniques checkpointing

Checkpointing following log: is done as follows:

Output all log records in memory to stable storage

Output to disk all modified buffer blocks

Output to log on stable storage a < checkpoint L> record.

Transactions are not allowed to perform any actions while checkpointing is in progress.

Fuzzy checkpointing allows transactions to progress while the most time consuming parts of checkpointing are in progress

Performed as described on next slide

Advanced Recovery Techniques:Checkpointing


Advanced recovery techniques fuzzycheckpointing

Fuzzy checkpointing following log: is done as follows:

Temporarily stop all updates by transactions

Write a <checkpointL> log record and force log to stable storage

Note list M of modified buffer blocks

Now permit transactions to proceed with their actions

Output to disk all modified buffer blocks in list M

blocks should not be updated while being output

Follow WAL: all log records pertaining to a block must be output before the block is output

Store a pointer to the checkpoint record in a fixed position last_checkpoint on disk

When recovering using a fuzzy checkpoint, start scan from the checkpoint record pointed to by last_checkpoint

Log records before last_checkpoint have their updates reflected in database on disk, and need not be redone.

Incomplete checkpoints, where system had crashed while performing checkpoint, are handled safely

Advanced Recovery Techniques:Fuzzycheckpointing


Normal checkpointing example

buffer following log:

M1

A

M2

B

Normal Checkpointing Example

Assume Last_chkpoint L1

(4) End chkpoint L2

(5) Set Last_chkpoint L2

(6) Resume the stopped transactions

(1) Stop all transactions

(2) Begin chkpoint L2

  • (3) during chkpoint

  • Log force

  • Flush buffer


Fuzzy checkpointing example

buffer following log:

M1

A

M2

B

Fuzzy Checkpointing Example

Assume Last_chkpoint L1

(4) End chkpoint L2

(5) Resume the stopped transactions

(1) Stop all transactions

(2) Begin chkpoint L2

(7) Set Last_chkpoint L2

(6) Flush buffer

  • (3) during chkpoint

  • Log force


Chapter 17 recovery system10

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery With Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Aries
ARIES following log:

  • ARIES is a “state of the art” recovery method

    • Incorporates numerous optimizations to reduce overheads during normal processing and to speed up recovery

    • The “advanced recovery algorithm” we studied earlier is modeled after ARIES, but greatly simplified by removing optimizations

  • Unlike the advanced recovery algorithm, ARIES has

    • Log sequence number (LSN) to identify log records

      • Stores LSNs in pages to identify what updates have already been applied to a database page

    • Physiological redo

    • Dirty page table to avoid unnecessary redos during recovery

    • Fuzzy checkpointing that only records information about dirty pages, and does not require dirty pages to be written out at checkpoint time

      • More coming up on each of the above …


Aries optimizations
ARIES Optimizations following log:

  • Physiological redo

    • Affected page is physically identified, action within page can be logical

      • Used to reduce logging overheads

        • e.g. when a record is deleted and all other records have to be moved to fill hole

          • Physiological redo can log just the record deletion

          • Physical redo would require logging of old and new values for much of the page

      • Requires page to be output to disk atomically

        • Easy to achieve with hardware RAID, also supported by some disk systems

        • Incomplete page output can be detected by checksum techniques,

          • But extra actions are required for recovery

          • Treated as a media failure


Aries data structures
ARIES Data Structures following log:

  • Log sequence number (LSN) identifies each log record

    • Must be sequentially increasing

    • Typically an offset from beginning of log file to allow fast access

      • Easily extended to handle multiple log files

  • Each page contains a PageLSN which is the LSN of the last log record whose effects are reflected on the page

    • To update a page:

      • X-latch the pag, and write the log record

      • Update the page

      • Record the LSN of the log record in PageLSN

      • Unlock page

    • Page flush to disk S-latches page

      • Thus page state on disk is operation consistent

        • Required to support physiological redo

    • PageLSN is used during recovery to prevent repeated redo

      • Thus ensuring idempotence


Aries data structures cont
ARIES Data Structures (Cont.) following log:

  • Each log record contains LSN of previous log record of the same transaction

    • LSN in log record may be implicit

  • Special redo-only log record called compensation log record (CLR) used to log actions taken during recovery that never need to be undone

    • Also serve the role of operation-abort log records used in advanced recovery algorithm

    • Have a field UndoNextLSN to note next (earlier) record to be undone

      • Records in between would have already been undone

      • Required to avoid repeated undo of already undone actions

LSN TransId PrevLSN RedoInfo UndoInfo

LSN TransID UndoNextLSN RedoInfo


Aries data structures cont1
ARIES Data Structures (Cont.) following log:

  • DirtyPageTable

    • List of pages in the buffer that have been updated

    • Contains, for each such page

      • PageLSN of the page

      • RecLSN is an LSN such that log records before this LSN have already been applied to the page version on disk

        • Set to current end of log when a page is inserted into dirty page table (just before being updated)

        • Recorded in checkpoints, helps to minimize redo work

  • Checkpoint log record

    • Contains:

      • DirtyPageTable and list of active transactions

      • For each active transaction, LastLSN, the LSN of the last log record written by the transaction

    • Fixed position on disk notes LSN of last completed checkpoint log record


Aries running example 1

T1 write page 1 following log:

T2 write page 2

T1 write page 1

T3 write page 4

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (1)

Active Transaction List

Dirty Page Table


Aries running example 2

T1 write page 1 following log:< 1, T1, - >

T2 write page 2

T1 write page 1

T3 write page 4

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (2)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 3

T1 write page 1 following log:< 1, T1, - >

T2 write page 2 < 2, T2, - >

T1 write page 1

T3 write page 4

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (3)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 4

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1 < 3, T1, 1 >

T3 write page 4

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (4)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 5

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4< 4, T3, - >

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (5)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 6

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (6)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 7

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits < 5, T2 commit >

Begin Checkpoint

End Checkpoint

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (7)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 8

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint < begin chkpt >

End Checkpoint < end chkpt >

T4 write page 3

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (8)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 9

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3< 8, T4, - >

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (9)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 10

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

(Page 4 flushed to disk)

T3 write page 2

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (10)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 11

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

(Page 4 flushed to disk)

T3 write page 2 < 9, T3, 4 >

T3 commits

T1 writes page 4

Crash!

ARIES Running Example (11)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 12

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

(Page 4 flushed to disk)

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4

Crash!

ARIES Running Example (12)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries running example 13

T1 write page 1 following log:< 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

(Page 1 flushed to disk)

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

(Page 4 flushed to disk)

T3 write page 2 < 9, T3, 4 >

T3 commits< 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Crash!

ARIES Running Example (13)

RedoInfo UndoInfo <LSN TransId PrevLSN>

Active Transaction List

Dirty Page Table


Aries recovery algorithm
ARIES Recovery Algorithm following log:

ARIES recovery involves three passes

  • Analysis pass: Determines

    • Which transactions to undo

    • Which pages were dirty (disk version not up to date) at time of crash

    • RedoLSN: LSN from which redo should start

  • Redo pass:

    • Repeats history, redoing all actions from RedoLSN

      • RecLSN and PageLSNs are used to avoid redoing actions already reflected on page

  • Undo pass:

    • Rolls back all incomplete transactions

      • Transactions whose abort was complete earlier are not undone

        • Key idea: no need to undo these transactions: earlier undo actions were logged, and are redone as required


Aries recovery analysis
ARIES Recovery: Analysis following log:

Analysis pass

  • Starts from last complete checkpoint log record

    • Reads in DirtyPageTable from log record

    • Sets RedoLSN = min of RecLSNs of all pages in DirtyPageTable

      • In case no pages are dirty, RedoLSN = checkpoint record’s LSN

    • Sets undo-list = list of transactions in checkpoint log record

    • Reads LSN of last log record for each transaction in undo-list from checkpoint log record


Aries recovery analysis cont
ARIES Recovery: Analysis (Cont.) following log:

Analysis pass (cont.)

  • Scans forward from checkpoint

    • If any log record found for transaction not in undo-list, adds transaction to undo-list

    • Whenever an update log record is found

      • If page is not in DirtyPageTable, it is added with RecLSN set to LSN of the update log record

    • If transaction end log record found, delete transaction from undo-list

    • Keeps track of last log record for each transaction in undo-list

      • May be needed for later undo

  • At end of analysis pass:

    • RedoLSN determines where to start redo pass

    • RecLSN for each page in DirtyPageTable used to minimize redo work

    • All transactions in undo-list need to be rolled back


Aries recovery analysis pass 1
ARIES Recovery following log:– Analysis Pass (1)

Active Transaction List

from the last checkpoint

RedoInfo UndoInfo <LSN TransId PrevLSN>

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits< 10, T3 commit >

T1 writes page 4< 11, T1, 3 >

Dirty Page Table

from the last checkpoint

  • RedoLSN = min(1, 3) = 1

  • Undo-list = {T1, T3}


Aries recovery analysis pass 2
ARIES Recovery following log:– Analysis Pass (2)

Active Transaction List

from the last checkpoint

RedoInfo UndoInfo <LSN TransId PrevLSN>

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits< 10, T3 commit >

T1 writes page 4< 11, T1, 3 >

Dirty Page Table

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

  • Undo-list = {T1, T3, T4}


Aries recovery analysis pass 3
ARIES Recovery following log:– Analysis Pass (3)

Active Transaction List

from the last checkpoint

RedoInfo UndoInfo <LSN TransId PrevLSN>

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits< 10, T3 commit >

T1 writes page 4< 11, T1, 3 >

Dirty Page Table

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

  • Undo-list = {T1, T3, T4}


Aries recovery analysis pass 4
ARIES Recovery following log:– Analysis Pass (4)

Active Transaction List

from the last checkpoint

RedoInfo UndoInfo <LSN TransId PrevLSN>

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4< 11, T1, 3 >

Dirty Page Table

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

  • Undo-list = {T1, T3, T4}


Aries recovery analysis pass 5
ARIES Recovery following log:– Analysis Pass (5)

Active Transaction List

from the last checkpoint

RedoInfo UndoInfo <LSN TransId PrevLSN>

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Dirty Page Table

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

  • Undo-list = {T1, T4}


Aries redo pass
ARIES Redo Pass following log:

Redo Pass: Repeats history by replaying every action not already reflected in the page on disk, as follows:

  • Scans forward from RedoLSN.

    Whenever an update log record is found:

    • If the page is not in DirtyPageTable or the LSN of the log record is less than the RecLSN of the page in DirtyPageTable, then skip the log record

    • Otherwise fetch the page from disk. If the PageLSN of the page fetched from disk is less than the LSN of the log record, redo the log record

      NOTE: if either test is negative the effects of the log record have already appeared on the page. First test avoids even fetching the page from disk!


Aries recovery redo pass 1
ARIES Recovery following log:– Redo Pass (1)

Active Transaction List

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Dirty Page Table

from the last checkpoint

1 : No redo

page 1 is not in dirty page table


Aries recovery redo pass 2
ARIES Recovery following log:– Redo Pass (2)

Active Transaction List

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2 < 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Dirty Page Table

from the last checkpoint

2 : LSN 2 >= RecLSN 1  read page 2

PageLSN of the fetched page 0 < 2,

thus redo


Aries recovery redo pass 3
ARIES Recovery following log:– Redo Pass (3)

Active Transaction List

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2 < 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Dirty Page Table

from the last checkpoint

3 : No redo

4 : LSN 4 >= RecLSN 3  read page 4

PageLSN of the fetched page 4 >= 4,

thus no redo


Aries recovery redo pass 4
ARIES Recovery following log:– Redo Pass (4)

Active Transaction List

from the last checkpoint

  • RedoLSN = min(1, 3, 8) = 1

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2 < 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

Dirty Page Table

from the last checkpoint

8, 9, 11 : Redo


Aries undo actions

6' following log:

3'

1

4

5

2

3

4'

6

5'

1'

2'

ARIES Undo Actions

  • When an undo is performed for an update log record

    • Generate a CLR containing the undo action performed (actions performed during undo are logged physicaly or physiologically).

      • CLR for record n noted as n’ in figure below

    • Set UndoNextLSN of the CLR to the PrevLSN value of the update log record

      • Arrows indicate UndoNextLSN value

  • ARIES supports partial rollback

    • Used e.g. to handle deadlocks by rolling back just enough to release reqd. locks

    • Figure indicates forward actions after partial rollbacks

      • records 3 and 4 initially, later 5 and 6, then full rollback


Aries undo pass
ARIES: Undo Pass following log:

Undo pass

  • Performs backward scan on log undoing all transaction in undo-list

    • Backward scan optimized by skipping unneeded log records as follows:

      • Next LSN to be undone for each transaction set to LSN of last log record for transaction found by analysis pass.

      • At each step pick largest of these LSNs to undo, skip back to it and undo it

      • After undoing a log record

        • For ordinary log records, set next LSN to be undone for transaction to PrevLSN noted in the log record

        • For compensation log records (CLRs) set next LSN to be undo to UndoNextLSN noted in the log record

          • All intervening records are skipped since they would have been undo already

  • Undos performed as described earlier


Aries recovery undo pass 1
ARIES Recovery following log:– Undo Pass (1)

LSN TransID UndoNextLSN RedoInfo

Active Transaction List

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

CLR < 11’, T1, 3 >

  • Undo-list = {T1, T4}

Next record to undo

= max(3, 8) = 11

Last LSN T1

= prevLSN of record 11 = 3


Aries recovery undo pass 2
ARIES Recovery following log:– Undo Pass (2)

LSN TransID UndoNextLSN RedoInfo

Active Transaction List

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

CLR < 11’, T1, 3 >

CLR < 8’ T4, - >

  • Undo-list = {T1, T4}

Next record to undo

= max(3, 8) = 8

Last LSN T4

= prevLSN of record 8 = “-”

 Remove T4 from undo-list


Aries recovery undo pass 3
ARIES Recovery following log:– Undo Pass (3)

LSN TransID UndoNextLSN RedoInfo

Active Transaction List

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1 < 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

CLR < 11’, T1, 3 >

CLR < 8’ T4, - >

CLR < 3’, T1, 1 >

  • Undo-list = {T1}

Next record to undo = 3

Last LSN T1

= prevLSN of record 3 = 1


Aries recovery undo pass 4
ARIES Recovery following log:– Undo Pass (4)

LSN TransID UndoNextLSN RedoInfo

Active Transaction List

Log at crash

T1 write page 1 < 1, T1, - >

T2 write page 2< 2, T2, - >

T1 write page 1< 3, T1, 1 >

T3 write page 4 < 4, T3, - >

T2 commits< 5, T2 commit >

Begin Checkpoint< begin chkpt >

End Checkpoint< end chkpt >

T4 write page 3 < 8, T4, - >

T3 write page 2 < 9, T3, 4 >

T3 commits < 10, T3 commit >

T1 writes page 4 < 11, T1, 3 >

CLR < 11’, T1, 3 >

CLR < 8’ T4, - >

CLR < 3’, T1, 1 >

CLR < 1’, T1, - >

  • Undo-list = {T1}

Last LSN T1

= prevLSN of record 1 = “-”

 Remove T1 from undo-list

Empty undo-list

 Undo complete


Other aries features
Other ARIES Features following log:

  • Recovery Independence

    • Pages can be recovered independently of others

      • E.g. if some disk pages fail they can be recovered from a backup while other pages are being used

  • Savepoints:

    • Transactions can record savepoints and roll back to a savepoint

      • Useful for complex transactions

      • Also used to rollback just enough to release locks on deadlock


Other aries features cont
Other ARIES Features (Cont.) following log:

  • Fine-grained locking:

    • Index concurrency algorithms that permit tuple level locking on indices can be used

      • These require logical undo, rather than physical undo, as in advanced recovery algorithm

  • Recovery optimizations:

    For example:

    • Dirty page table can be used to prefetch pages during redo

    • Out of order redo is possible:

      • redo can be postponed on a page being fetched from disk, and performed when page is fetched.

      • Meanwhile other log records can continue to be processed


Chapter 17 recovery system11

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery With Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Remote backup systems

Remote backup systems provide following log:high availability by allowing transaction processing to continue even if the primary site is destroyed.

Primary site and secondary site should be synchronized

by sending all log records from primary to secondary

Remote Backup Systems


Remote backup systems cont

Detection of failure following log:: Backup site must detect when primary site has failed

To distinguish primary site failure from link failure, several communication links between the primary and the remote backup need to be maintained.

Transfer of control:

To take over control: the backup site first perform recovery using its copy of the database and all the long records it has received from the primary.

Thus, completed transactions are redone and incomplete transactions are rolled back.

When the backup site takes over processing it becomes the new primary

To transfer control back: to the old primary when it recovers, the old primary must receive redo logs from the old backup and apply all updates locally.

Remote Backup Systems (Cont.)


Remote backup systems cont1

Time to recover following log:: To reduce delay in takeover, the backup site periodically processes the redo log records (in effect, performing recovery from previous database state), performs a checkpoint, and can then delete earlier parts of the log.

Hot-Spare configuration permits very fast takeover:

The backup site continually processes redo log record as they arrive, applying the updates locally.

When failure of the primary is detected, the backup site rolls back incomplete transactions, and is immediately ready to process new transactions.

Alternative to the remote backup site: distributed database with replicated data

Transactions are required to update all replicas of any data item that they update

Transactions are accepted in any database replicas

Remote backup is faster and cheaper, but less tolerant to failure

more on this in Chapter 19

Spare: (동) 용서하다, 아끼다 (형) 예비의, 여벌의

Remote Backup Systems (Cont.)


Hot spare configuration example
Hot Spare Configuration following log:Example

LA : <T0, A, 950, 1000>

(a) Non-Hot spare configuration

backup

Process redo log records

when failure of the primary is detected

primary

network

Not immediately

processed

log

records

LA

A=1000

log

records

LA

A=950

(b) Hot spare configuration

backup

primary

Continually process

redo log record

as they arrive

network

immediately

processed

LA

A=950

log

records

LA

A=950

log

records


Remote backup systems cont2

Time to Commit following log:

Ensure durability of updates by delaying transaction commit until update is logged at backup

Avoid this delay by permitting lower degrees of durability.

One-safe:

commit as soon as transaction’s commit log record is written at primary

Problem: updates may not arrive at backup before it takes over.

Two-very-safe:

commit when transaction’s commit log record is written at primary and backup

Reduces availability since transactions cannot commit if either site fails.

Two-safe:

proceed as in two-very-safe if both primary and backup are active.

If only the primary is active, the transaction commits as soon as its commit log record is written at the primary.

Better availability than two-very-safe; avoids problem of lost transactions in one-safe.

Remote Backup Systems (Cont.)


Time to commit example
Time to Commit Example following log:

(a) One-safe; T1 can commiteven if L1 is not written in the log of the backup

Backup

Primary

T1 commit

network

L1 : <T1 commit>

log

records

L1

log

records

(b) Two-very-safe; T1 cannot commit if L1 is not written in the log of the backup

Backup

Primary

T1 commit

network

log

records

log

records

L1


Chapter 17 recovery system12

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery With Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


Ch 17 summary 1
Ch.17 Summary (1) following log:

  • A computer system, like any other mechanical or electrical device, is subject to failure.

    • There are a variety of causes of such failure, including disk crash, power failure, and software errors.

    • In each of these case, information concerning the database system is lost.

  • In addition to system failures, transactions may also fail for various reasons, such as violation of integrity constraints or deadlocks.

  • An integral part of a databases system is a recovery scheme that is responsible for the detection of failures and for the restoration of the database to a state that existed before the occurrence of the failure.

  • The various types of storage in a computer are volatile storage, nonvolatile storage, and stable storage.

    • Data in volatile storage, such as in RAM, are lost when the computer crashes.

    • Data in nonvolatile storage, such as disk, are not lost when the computer crashes, but may occasionally be lost because of failures such as disk crashes.

    • Data in stable storage are never lost.


Ch17 summary 2
Ch17. Summary (2) following log:

  • Stable storage that must be accessible online is approximated with mirrored disks, or other forms of RAID, which provide redundant data storage.

    • Offline, or archival, stable storage may consist of multiple tape copies of data stored in a physically secure location.

  • In case of failure, the state of the database system may no longer be consistent; that is, it may not reflect a state of the world that the database is supposed to capture.

    • To preserve consistency, we require that each transaction be atomic.

    • It is the responsibility of the recovery scheme to ensure the atomicity and durability property.

    • There are basically two different approaches for ensuring atomicity: log-base schemes and shadow paging.


Ch17 summary 3
Ch17. Summary (3) following log:

  • In log-based schemes, all updates are recorded on a log, which must be kept in stable storage.

    • In the deferred- modifications scheme, during the execution of a transaction, all the write operations are are deferred until the transaction partially commits, at which time the system sues the information on the log associated with the transaction in execution the deferred writes.

    • In the immediate-modifications scheme, the system applies all updates directly to the database. If a crash occurs, the system uses the information in the restoring the state of the system to a previous consistent state.

      To reduce the overhead of searching the log and redoing transactions, we can use the checkpointing technique.


Ch17 summary 4
Ch17. Summary (4) following log:

  • In shadow paging, two page tables are maintained during the life of a transaction: the current page table and the shadow page table.

    • When the transaction starts, both page tables are identical.

    • The shadow page table and page it points to are never changed during the duration of the transaction.

    • When the transaction partially commits, the shadow page table is discarded, and the current table becomes the new page table.

    • If the transaction aborts, the current page table is simply discarded.

  • If multiple transactions are allowed to execute concurrently, then the shadow-paging technique is not applicable, but the log-based technique can be used.

    • No transactions can allowed to update a data item that has already been updated by an incomplete transaction.

    • We an use strict two- phase locking to ensure this condition.


Ch17 summary 5
Ch17. Summary (5) following log:

  • Transaction processing is based on a storage model in which main memory holds a log buffer, a database buffer, and a system buffer.

    • The system buffer holds pages of system object code and local work areas of transaction.

  • Efficient implementation of a recovery scheme requires that the number of writes to the database and to stable storage be minimized.

    • Log records may be kept in volatile log buffer initially, but must be written to stable storage when one of the following conditions occurs:

    • Before the <Ti commit> log record may be output to stable storage, all log records pertaining to transaction must have been output to the database (in nonvolatile storage), all log records pertaining to data in that block must have been output to stable storage.

    • Before a block of data in main memory is output to the database (in nonvolatile storage), all log records pertaining to data in that block must have been output to stable storage


Ch17 summary 7
Ch17. Summary (7) following log:

  • To recover from failures that result in the loss of nonvolatile storage, we must dump the entire contents of the database onto stable storage periodically- say, once per day.

    • If a failure occurs that results in the loss of physical database blocks, we use the most recent dump in restoring the database to previous consistent state.

    • Once this restoration has been accomplished, we use the log to bring the databases system to the most recent consistent state.

  • Advanced recovery techniques support high-concurrency locking techniques, such as those used for B+ tree concurrency control. These techniques are based on logical (operation) undo, and follow the principle of repeating history.

    • When recovering from system failure, the system performs a redo pass using the log, followed by an undo pass on the log to roll back incomplete transactions.


Ch18 summary 8
Ch18. Summary (8) following log:

  • The ARIES recovery scheme is a state-of-the-art scheme that supports a number of features to provide greater concurrency, reduce logging overheads, and minimize recovery time.

    • It is also based on repeating of history, and allows logical undo operations.

    • The scheme flushes pages on a continuous basis and does not need to flush all pages at the time of a checkpoint.

    • It uses log sequence numbers (LSNs) to implement a variety of optimizations that reduce the time taken for recovery.

  • Remote backup systems provide a high degree of availability, allowing transaction processing to continue even if the primary site is destroyed by a fire, flood, or earthquake.


Ch17 bibliographical notes 1
Ch17. Bibliographical Notes (1) following log:

  • A comprehensive presentation of the principles of recovery is offered by Hearder and Reuter [1983].

  • Gray and Reuter [1993] is an excellent textbook source of information about recovery, including interesting implementation and historical details.

  • Bernstein et al. [1987] is an early textbook source of information on concurrency control and recovery.

  • Two early papers that present initial theoretical work in the area of recovery are Davies[1973] and Bjork[1973].

  • An overview of the recovery scheme of System R is presented by Gray et al. [1981b].

  • The shadow-paging mechanism of System R is described by Lorie [1977].

  • Tutorial and survey papers on various recovery techniques for database systems include Gray [1978], Lindsay et al [1980], and Verhofstad [1978].

  • The concepts of fuzzy checkpointing and fuzzy dumps are described in Lindsay et al [1980].


Ch17 bibliographical notes 2
Ch17. Bibliographical Notes (2) following log:

  • Chandy et al. [1975], which describes analytic models for rollback and recovery strategies in database systems, is another early work in this area.

  • The state of the art in recovery methods is best illustrated by the ARIES recovery method, described in Mohan et al. [1992] and Mohan [1990b].

  • Aries and its variants are used in several databases products, including IBM DB2 and Mircrosoft SQL Server.

  • Recovery in Oracle is described in Lahiri et al. [2001].

  • Specialized recovery techniques for index structures are described in Mohan and Levine [1992] and Mohan [1993]; Mohan and Narang [1994] describes recovery techniques for client-server architectures, while Mohan and Narang [1991] and Mohan and Narang [1992] describe recovery techniques for parallel databases architectures.

  • Remote backup for disaster recovery (loss of an entire computing facility by, for example. Fire, flood, or earthquake) is considered in King et al. [1991] and Polyzois and Garcia-Molina [1994].

  • Chapter 24 lists references pertaining to long-duration transactions and related recovery issues.


Chapter 17 recovery system13

17.1 Failure Classification following log:

17.2 Storage Structure

17.3 Recovery and Atomicity

17.4 Log-Based Recovery

Aux: Shadow Paging

17.5 Recovery With Concurrent Transactions

17.6 Buffer Management

17.7 Failure with Loss of Nonvolatile Storage

17.8 Advanced Recovery Techniques

Aux: ARIES Recovery Algorithm

17.9 Remote Backup Systems

19.10 Summary

Chapter 17: Recovery System


End of chapter

End of Chapter following log:


ad