Database , 13 Replication

Distributed DBMS ©M. T. Özsu & P. Valduriez Ch.13/1
Outline
• Introduction
• Background
• Distributed Database Design
• Database Integration
• Semantic Data Control
• Distributed Query Processing
• Distributed Transaction Management
• Data Replication
➡ Consistency criteria
➡ Replication protocols
➡ Replication and failure management
• Parallel Database Systems
• Distributed Object DBMS
• Peer-to-Peer Data Management
• Web Data Management
• Current Issues

Replication
• Why replicate?
➡ System availability
✦ Avoid single points of failure
➡ Performance
✦ Localization
➡ Scalability
✦ Scalability in numbers and geographic area
➡ Application requirements
• Why not replicate?
➡ Replication transparency
➡ Consistency issues
✦ Updates are costly
✦ Availability may suffer if not careful

Execution Model
• There are physical copies of logical objects in the system.
• Operations are specified on logical objects, but translated to operate on
physical objects.
• One-copy equivalence
➡ The effect of transactions performed by clients on replicated objects should be
the same as if they had been performed on a single set of objects.
x
x1 x2 xn
…
Physical data item (replicas, copies)
Logical data item
Write(x)
Write(x1) Write(x2) Write(xn)

Replication Issues
• Consistency models - how do we reason about the consistency of the
“global execution state”?
➡ Mutual consistency
➡ Transactional consistency
• Where are updates allowed?
➡ Centralized
➡ Distributed
• Update propagation techniques – how do we propagate updates to one
copy to the other copies?
➡ Eager
➡ Lazy

Consistency
• Mutual Consistency
➡ How do we keep the values of physical copies of a logical data item
synchronized?
➡ Strong consistency
✦ All copies are updated within the context of the update transaction
✦ When the update transaction completes, all copies have the same value
✦ Typically achieved through 2PC
➡ Weak consistency
✦ Eventual consistency: the copies are not identical when update transaction
completes, but they eventually converge to the same value
✦ Many versions possible:
✓ Time-bounds
✓ Value-bounds
✓ Drifts

Transactional Consistency
• How can we guarantee that the global execution history over replicated
data is serializable?
• One-copy serializability (1SR)
➡ The effect of transactions performed by clients on replicated objects should be
the same as if they had been performed one at-a-time on a single set of objects.
• Weaker forms are possible
➡ Snapshot isolation
➡ RC-serializability

Example 1
Site A Site B Site C
x x, y x, y, z
T1: x ← 20 T2: Read(x) T3: Read(x)
Write(x) x ← x+y Read(y)
Commit Write(y) z ← (x∗y)/100
Commit Write(z)
Commit
Consider the three histories:
HA={W1(xA), C1}
HB={W1(xB), C1, R2(xB), W2(yB), C2}
HC={W2(yC), C2, R3(xC), R3(yC),W3(zC), C3, W1(xC),C1}
Global history non-serializable: HB: T1→T2, HC: T2→T3→T1
Mutually consistent: Assume xA=xB=xC=10, yB=yC=15,yC=7 to begin; in the
end xA=xB=xC=20, yB=yC=35,yC=3.5

Example 2
Site A Site B
x x
T1: Read(x) T2: Read(x)
x ← x+5 x ← x∗10
Write(x) Write(x)
Commit Commit
Consider the two histories:
HA={R1(xA),W1(xA), C1, W2(xA), C2}
HB={R1(xB), W2(xB), C2, W1(xB), C1}
Global history non-serializable: HA: T1→ T2, HB: T2→ T1
Mutually inconsistent: Assume xA=xB=1 to begin; in the end xA=10,
xB=6

Update Management Strategies
• Depending on when the updates are propagated
➡ Eager
➡ Lazy
• Depending on where the updates can take place
➡ Centralized
➡ Distributed
Eager
Lazy
Centralized Distributed

Eager Replication
• Changes are propagated within the scope of the transaction making the
changes. The ACID properties apply to all copy updates.
➡ Synchronous
➡ Deferred
• ROWA protocol: Read-one/Write-all
Site 1 Site 2 Site 3 Site 4
Transaction
updates commit




Lazy Replication
●Lazy replication first executes the updating transaction on one copy. After
the transaction commits, the changes are propagated to all other copies
(refresh transactions)
●While the propagation takes place, the copies are mutually inconsistent.
●The time the copies are mutually inconsistent is an adjustable parameter
which is application dependent.
Transaction
updates commit
 


Centralized
●There is only one copy which can be updated (the master), all others (slave
copies) are updated reflecting the changes to the master.

Distributed
●Changes can be initiated at any of the copies. That is, any of the sites which
owns a copy can update the value of the data item.
Transaction
updates commit
Transaction
updates commit

Forms of Replication
Eager
+ No inconsistencies (identical copies)
+ Reading the local copy yields the most
up to date value
+ Changes are atomic
− A transaction has to update all sites
− Longer execution time
− Lower availability
Lazy
+ A transaction is always local (good
response time)
− Data inconsistencies
− A local read does not always return the
most up-to-date value
− Changes to all copies are not
guaranteed
− Replication is not transparent
Centralized
+ No inter-site synchronization is
necessary (it takes place at the
master)
+ There is always one site which has
all the updates
− The load at the master can be high
− Reading the local copy may not
yield the most up-to-date value
Distributed
+ Any site can run a transaction
+ Load is evenly distributed
− Copies need to be synchronized

Replication Protocols
Eager
Lazy
Eager centralized Eager distributed
Lazy distributedLazy centralized
The previous ideas can be combined into 4 different replication protocols:

Eager Centralized Protocols
• Design parameters:
➡ Distribution of master
✦ Single master: one master for all data items
✦ Primary copy: different masters for different (sets of) data items
➡ Level of transparency
✦ Limited: applications and users need to know who the master is
✓ Update transactions are submitted directly to the master
✓ Reads can occur on slaves
✦ Full: applications and users can submit anywhere and the operations will be
forwarded to the master
✓ Operation-based forwarding
• Four alternative implementation architectures, only three are meaningful:
➡ Single master, limited transparency
➡ Single master, full transparency
➡ Primary copy, full transparency

Eager Single Master/Limited
Transparency
• Applications submit update transactions directly to the master
• Master:
➡ Upon read: read locally and return to user
➡ Upon write: write locally, multicast write to other replicas (in FFO
timestamps order)
➡ Upon commit request: run 2PC coordinator to ensure that all have really
installed the changes
➡ Upon abort: abort and inform other sites about abort
• Slaves install writes that arrive from the master

Eager Single Master/Limited
Transparency (cont’d)
• Applications submit read transactions directly to an appropriate slave
• Slave
➡ Upon read: read locally
➡ Upon write from master copy: execute conflicting writes in the proper order
(FIFO or timestamp)
➡ Upon write from client: refuse (abort transaction; there is error)
➡ Upon commit request from read-only: commit locally
➡ Participant of 2PC for update transaction running on primary

Eager Single Master/
Full Transparency
Coordinating TM
1. Send op(x) to the master site
2. Send Read(x) to any site that has x
3. Send Write(x) to all the slaves
where a copy of x exists
4. When Commit arrives, act as
coordinator for 2PC
Master Site
1. If op(x) = Read(x): set read lock on
x and send “lock granted” msg to
the coordinating TM
2. If op(x) = Write(x)
1. Set write lock on x
2. Update local copy of x
3. Inform coordinating TM
3. Act as participant in 2PC
Applications submit all transactions to the Transaction Manager at their
own sites (Coordinating TM)

Eager Primary Copy/Full
Transparency
• Applications submit transactions directly to their local TMs
• Local TM:
➡ Forward each operation to the primary copy of the data item
➡ Upon granting of locks, submit Read to any slave, Write to all slaves
➡ Coordinate 2PC

Eager Primary Copy/Full
Transparency (cont’d)
• Primary copy site
➡ Read(x): lock xand reply to TM
➡ Write(x): lock x, perform update, inform TM
➡ Participate in 2PC
• Slaves: as before

Eager Distributed Protocol
• Updates originate at any copy
➡ Each sites uses 2 phase locking.
➡ Read operations are performed locally.
➡ Write operations are performed at all sites (using a distributed locking
protocol).
➡ Coordinate 2PC
• Slaves:
➡ As before

Eager Distributed Protocol
(cont’d)
• Critical issue:
➡ Concurrent Writes initiated at different master sites are executed in the same
order at each slave site
➡ Local histories are serializable (this is easy)
• Advantages
➡ Simple and easy to implement
• Disadvantage
➡ Very high communication overhead
✦ n replicas; m update operations in each transaction: n*m messages (assume no
multicasting)
✦ For throughput of k tps: k* n*m messages
• Alternative
➡ Use group communication + deferred update to slaves to reduce messages

Lazy Single Master/Limited
Transparency
• Update transactions submitted to master
• Master:
➡ Upon write: write locally and return to user
➡ Upon commit/abort: terminate locally
➡ Sometime after commit: multicast updates to slaves (in order)
• Slaves:
➡ Upon read: read locally
➡ Refresh transactions: install updates

Lazy Primary Copy/Limited
Transparency
• There are multiple masters; each master execution is similar to lazy single
master in the way it handles transactions
• Slave execution complicated: refresh transactions from multiple masters
and need to be ordered properly

Lazy Primary Copy/Limited
Transparency – Slaves
• Assign system-wide unique timestamps to refresh transactions and execute
them in timestamp order
➡ May cause too many aborts
• Replication graph
➡ Similar to serialization graph, but nodes are transactions (T) + sites (S); edge
〈Ti,Sj〉exists iff Ti performs a Write(x) and x is stored in Sj
➡ For each operation (opk), enter the appropriate nodes (Tk) and edges; if graph
has no cycles, no problem
➡ If cycle exists and the transactions in the cycle have been committed at their
masters, but their refresh transactions have not yet committed at slaves, abort
Tk; if they have not yet committed at their masters, Tkwaits.
• Use group communication

Lazy Single Master/Full
Transparency
• This is very tricky
➡ Forwarding operations to a master and then getting refresh transactions
cause difficulties
• Two problems:
➡ Violation of 1SR behavior
➡ A transaction may not see its own reads
• Problem arises in primary copy/full transparency as well

Example 3
Site M (Master) holds x, y; SiteB holds slave copies of x, y
T1: Read(x), Write(y), Commit
T2: Read(x), Write(y), Commit

Example 4
• Master site M holds x, site C holds slave copy of x
• T3: Write(x), Read(x), Commit
• Sequence of execution
1. W3(x) submitted at C, forwarded to M for execution
2. W3(x) is executed at M, confirmation sent back to C
3. R3(x) submitted at C and executed on the local copy
4. T3 submits Commit at C, forwarded to M for execution
5. M executes Commit, sends notification to C, which also commits T3
6. M sends refresh transaction for T3 to C (for W3(x) operation)
7. C executes the refresh transaction and commits it
• When C reads x at step 3, it does not see the effects of Write at step 2

Lazy Single Master/
Full Transparency - Solution
• Assume T = Write(x)
• At commit time of transaction T, the master generates a timestamp for it
[ts(T)]
• Master sets last_modified(xM) ← ts(T)
• When a refresh transaction arrives at a slave site i, it also sets
last_modified(xi) ← last_modified(xM)
• Timestamp generation rule at the master:
➡ ts(T) should be greater than all previously issued timestamps and should
be less than the last_modified timestamps of the data items it has accessed. If
such a timestamp cannot be generated, then T is aborted.

Lazy Distributed Replication
• Any site:
➡ Upon write: write locally and return to user
➡ Upon commit/abort: terminate locally
➡ Sometime after commit: send refresh transaction
➡ Upon message from other site
✦ Detect conflicts
✦ Install changes
✦ Reconciliation may be necessary

Reconciliation
• Such problems can be solved using pre-arranged patterns:
➡ Latest update win (newer updates preferred over old ones)
➡ Site priority (preference to updates from headquarters)
➡ Largest value (the larger transaction is preferred)
• Or using ad-hoc decision making procedures:
➡ Identify the changes and try to combine them
➡ Analyze the transactions and eliminate the non-important ones
➡ Implement your own priority schemas

Replication Strategies
+Updates do not need to be
coordinated
+No inconsistencies
- Longest response time
- Only useful with few updates
- Local copies are can only be
read
+No inconsistencies
+Elegant (symmetrical solution)
- Long response times
- Updates need to be
coordinated
+No coordination necessary
+Short response times
- Local copies are not up to date
- Inconsistencies
+No centralized coordination
+Shortest response times
- Inconsistencies
- Updates can be lost
(reconciliation)

Group Communication
• A node can multicast a message to all nodes of a group with a delivery
guarantee
• Multicast primitives
➡ There are a number of them
➡ Total ordered multicast: all messages sent by different nodes are delivered in
the same total order at all the nodes
• Used with deferred writes, can reduce communication overhead
➡ Remember eager distributed requires k*m messages (with multicast) for
throughput of ktps when there are n replicas and m update operations in each
transaction
➡ With group communication and deferred writes: 2k messages

Failures
• So far we have considered replication protocols in the absence of failures
• How to keep replica consistency when failures occur
➡ Site failures
✦ Read One Write All Available (ROWAA)
➡ Communication failures
✦ Quorums
➡ Network partitioning
✦ Quorums

ROWAA with Primary Site
• READ = read any copy, if time-out, read another copy.
• WRITE = send W(x) to all copies. If one site rejects the operation, then
abort. Otherwise, all sites not responding are “missing writes”.
• VALIDATION = To commit a transaction
➡ Check that all sites in “missing writes” are still down. If not, then abort the
transaction.
✦ There might be a site recovering concurrent with transaction updates and these
may be lost
➡ Check that all sites that were available are still available. If some do not
respond, then abort.

Distributed ROWAA
• Each site has a copy of V
➡ V represents the set of sites a site believes is available
➡ V(A) is the “view” a site has of the system configuration.
• The view of a transaction T [V(T)] is the view of its coordinating site, when the
transaction starts.
➡ Read any copy within V; update all copies in V
➡ If at the end of the transaction the view has changed, the transaction is aborted
• All sites must have the same view!
• To modify V, run a special atomic transaction at all sites.
➡ Take care that there are no concurrent views!
➡ Similar to commit protocol.
➡ Idea: Vs have version numbers; only accept new view if its version number is higher
than your current one
• Recovery: get missed updates from any active node
➡ Problem: no unique sequence of transactions

Quorum-Based Protocol
• Assign a vote to each copy of a replicated object (say Vi) such that ∑iVi = V
• Each operation has to obtain a read quorum (Vr) to read and a write
quorum (Vw) to write an object
• Then the following rules have to be obeyed in determining the quorums:
➡ Vr+ Vw>V an object is not read and written by two transactions
concurrently
➡ Vw>V/2 two write operations from two transactions cannot occur
concurrently on the same object

Quorum Example
Three examples of the voting algorithm:
a) A correct choice of read and write set
b) A choice that may lead to write-write conflicts
c) ROWA
From Tanenbaum and van Steen, Distributed Systems: Principles and Paradigms
© Prentice-Hall, Inc. 2002

Database , 13 Replication

Recommended

More Related Content

What's hot (20)

Similar to Database , 13 Replication (20)

More from Ali Usman (20)

Recently uploaded (20)

Database , 13 Replication