Download Object-Oriented & Object

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Final Exam Review
Last Lecture
R&G - All Chapters Covered
The end crowns all,
And that old common arbitrator, Time,
Will one day end it.
William Shakespeare.
Troilus and Cressida.
Topics Covered
• Transactions
• Concurrency Control
• Crash Recovery
ACID properties of Transaction Executions
•
A tomicity:
All actions in the Xact happen, or none
happen.
•
C onsistency:
If each Xact is consistent, and the DB
starts consistent, it ends up consistent.
• I solation:
Execution of one Xact is isolated from that
of other Xacts.
•
D urability:
If a Xact commits, its effects persist.
Concurrency Control
• Transaction: basic unit of operation
– made up of reads and writes
• Goal: ACID Transactions
• A & D are provided by Crash Recovery
• C & I are provided by Concurrency Control
• Bottom line: reads and writes for various
transactions MUST be ordered such that the
final state of the database is the same as some
serial ordering of the transactions
Scheduling Transactions
• Serial schedule: Schedule that does not interleave the
actions of different transactions.
• Equivalent schedules: For any database state, the
effect (on the set of objects in the database) of
executing the first schedule is identical to the effect of
executing the second schedule.
• Serializable schedule: A schedule that is equivalent to
some serial execution of the transactions.
(Note: If each transaction preserves consistency, every
serializable schedule preserves consistency. )
Conflict Serializable Schedules
• Two schedules are conflict equivalent if:
– Involve the same actions of the same transactions
– Every pair of conflicting actions is ordered the
same way
• Schedule S is conflict serializable if S is conflict
equivalent to some serial schedule
Example
• A schedule that is not conflict serializable:
T1:
T2:
R(A), W(A),
R(A), W(A), R(B), W(B)
R(B), W(B)
A
T1
T2
Dependency graph
B
• The cycle in the graph reveals the problem.
The output of T1 depends on T2, and viceversa.
Dependency Graph
• Dependency graph: One node per Xact; edge
from Ti to Tj if an operation of Ti conflicts with
an operation of Tj and Ti’s operation appears
earlier in the schedule than the conflicting
operation of Tj.
• Theorem: Schedule is conflict serializable if
and only if its dependency graph is acyclic
An Aside: View Serializability
• Schedules S1 and S2 are view equivalent if:
– If Ti reads initial value of A in S1, then Ti also reads initial value
of A in S2
– If Ti reads value of A written by Tj in S1, then Ti also reads
value of A written by Tj in S2
– If Ti writes final value of A in S1, then Ti also writes final value
of A in S2
T1: R(A)
W(A)
T2:
W(A)
T3:
W(A)
•
T1: R(A),W(A)
T2:
W(A)
T3:
W(A)
View serializability is “weaker” than conflict serializability!
– Every conflict serializable schedule is view serializable, but not
vice versa!
– I.e. admits more legal schedules
Approaches to Concurrency Control
• 2PL - all objects have Shared and eXclusive locks
– once one lock is released, no more locks may be acquired
– Strict 2PL: don’t release locks until commit time
– Conservative 2PL: acquire all locks at start, release all at end
• Locking issues
– must either prevent or detect deadlock
– may want multiple granularity locks (table, page, record)
using IS, IX, SIX, S, X locks (check compatibility matrix!)
– locking in B-trees usually not 2PL
– phantom problem: locking all records of a given criteria
(e.g., age > 20)
Multiple-Granularity Locks
• Hard to decide what granularity to lock (tuples
vs. pages vs. tables).
• Shouldn’t have to make same decision for all
transactions!
• Data “containers” are nested:
Database
contains
Tables
Pages
Tuples
Solution: New Lock Modes, Protocol
• Allow Xacts to lock at each level, but with a special
protocol using new “intention” locks:
• Still need S and X locks, but before locking an item,
Xact must have proper intension locks on all its
ancestors in the granularity hierarchy.
Database
Tables
Pages
Tuples
IS – Intent to get S lock(s) at
finer granularity.
 IX – Intent to get X lock(s)
at finer granularity.
 SIX mode: Like S & IX at
the same time. Why useful?
IS IX

SIX
IS   
IX  

S 
S

SIX
X

X
Multiple Granularity Lock Protocol
• Each Xact starts from the root of the
hierarchy.
• To get S or IS lock on a node, must hold IS or IX
on parent node.
– What if Xact holds SIX on parent? S on parent?
• To get X or IX or SIX on a node, must hold IX or
SIX on parent node.
• Must release locks in bottom-up order.
Protocol is correct in that it is equivalent to directly setting
locks at the leaf levels of the hierarchy.
Examples – 2 level hierarchy
• T1 scans R, and updates a few tuples:
– T1 gets an SIX lock on R, then get X lock on tuples that are
Tables
updated.
• T2 uses an index to read only part of R:
Tuples
– T2 gets an IS lock on R, and repeatedly gets an S lock on
tuples of R.
• T3 reads all of R:
IS IX SIX S X
– T3 gets an S lock on R.
IS    
– OR, T3 could behave like T2; can
IX  
use lock escalation to decide which.
SIX
• Lock escalation

S 
– Dynamically asks for coarser-grained locks

when too many low level locks acquired
X
Other Approaches to CC
• Optimistic CC
• Timestamp CC
• Multiversion CC
NOT COVERED in CLASS – NOT ON FINAL
Question from old final - True/False
• “Our lock manager uses S and X locks, which
guarantees that all schedules are conflictserializable”
• “Because we use 2PL, deadlocks can never
occur”
• “Because we use Strict 2PL, there will be no
cascading aborts”
Question from old final - True/False
• “Our lock manager uses S and X locks, which
guarantees that all schedules are conflictserializable”
• “Because we use 2PL, deadlocks can never
occur”
• “Because we use Strict 2PL, there will be no
cascading aborts”
More questions ..
• Read Uncommitted (UR):
– Always read data without setting any locks
– Always set exclusive locks before writing and hold
them until end-of-transaction.
• Or: No S-Locks, Strict 2-Phase X-Locks. Which of the
following conflicts can occur between a UR Xact, and
a Strict 2PL Xact:
– Write-Write
– Read-Write
– Write-Read
More questions ..
• Read Committed (RC):
– Obtain locks in the same way as Strict 2PL, and
release exclusive locks at the end-of-transaction
just like Strict 2PL.
• Or: RC transactions use Short-Term S-Locks, and
Strict 2-Phase X-Locks. Which of the following
conflicts can occur between a RC Xact, and a Strict
2PL Xact:
– Write-Write
– Read-Write
– Write-Read
More questions ..
• All SQL lock modes – even the relaxed modes – use
Strict 2-Phase Write Locks. Consider what would
happen if a transaction released a shared lock
immediately after a read, and released an exclusive
lock immediately after a write (Short-Term R-Locks
and Short-Term X-Locks). Which of the following
could happen in those cases?
– Write-Write conflicts.
– Inconsistent data in the database
– Deadlocks
Crash Recovery
•
•
•
•
ACID - need way to ensure A & D
We studied approach of Aries system
Buffer management Steal, no Force
Every Write to a page is first logged in WAS
– log record is in stable storage before data page on disk
– log record has Xact#, before value, after value
• Checkpoints record which pages dirty, which XActs
running
Buffer Mgmt Plays a Key Role
• Force policy – make sure that every update is on disk
before commit.
– Provides durability without REDO logging.
– But, can cause poor performance.
• No Steal policy – don’t allow buffer-pool frames with
uncommited updates to overwrite committed data on
disk.
– Useful for ensuring atomicity without UNDO logging.
– But can cause poor performance.
Of course, there are some nasty details for getting
Force/NoSteal to work…
Preferred Policy: Steal/No-Force
• This combination is most complicated but allows for
highest performance.
• NO FORCE (complicates enforcing Durability)
– What if system crashes before a modified page written by
a committed transaction makes it to disk?
– Write as little as possible, in a convenient place, at commit
time, to support REDOing modifications.
• STEAL (complicates enforcing Atomicity)
– What if the Xact that performed udpates aborts?
– What if system crashes before Xact is finished?
– Must remember the old value of P (to support UNDOing
the write to page P).
Buffer Management summary
No Steal
No Force
Force
Steal
Fastest
Slowest
Performance
Implications
No Steal
No Force No UNDO
REDO
Steal
UNDO
REDO
No UNDO UNDO
Force
No REDO No REDO
Logging/Recovery
Implications
Basic Idea: Logging
• Record REDO and UNDO information, for every
update, in a log.
– Sequential writes to log (put it on a separate disk).
– Minimal info (diff) written to log, so multiple updates
fit in a single log page.
• Log: An ordered list of REDO/UNDO actions
– Log record contains:
<XID, pageID, offset, length, old data, new data>
– and additional control info (which we’ll see soon).
Write-Ahead Logging (WAL)
• The Write-Ahead Logging Protocol:
 Must force the log record for an update before the
corresponding data page gets to disk.
 Must force all log records for a Xact before commit. (alt.
transaction is not committed until all of its log records including
its “commit” record are on the stable log.)
• #1 (with UNDO info) helps guarantee Atomicity.
• #2 (with REDO info) helps guarantee Durability.
• This allows us to implement Steal/No-Force
• Exactly how is logging (and recovery!) done?
– We’ll look at the ARIES algorithms from IBM.
Transaction Commit
•
•
•
•
write Commit record to log
flush log tail to stable storage
remove Xact from Xact table
write End record to log
Transaction Abort
• write Abort record to log
• go back through log, undoing each write (and
add CLR to log)
• when done, write End record to log
Crash Recovery - 3 phases
• Analysis: starting from checkpoint, go forward in the
log to see:
– what pages were dirty
– what transactions were active at time of crash
• Redo: start from oldest transaction that wrote to a
dirty page, and redo all writes to dirty pages.
• Undo: start at the end of the log (time of crash),
work backward undoing all writes made by
transactions that were active at time of crash
– What happens when you encounter a CLR ?