Chapter 4
Distributed Databases and Client-Server Architectures
Outline
Distributed Database Concepts
Data Fragmentation, Replication and Allocation
Types of Distributed Database Systems
Query Processing in distributed database systems
Concurrency Control and Recovery
Client-Server Architecture
1
Distributed Database Concepts
A distributed database (DDB) is a collection of multiple
logically related databases distributed over a computer
network
A distributed database management system (DDMS) is a
software system that manages a distributed database
while making the distribution transparent to the user
A transaction can be executed by multiple networked
computers in a unified manner
2
Distributed Database System
Advantages
Management of distributed data with different levels
of transparency:
This refers to the physical placement of data (files,
relations, etc.) which is not known to the user
(distribution transparency).
3
Distributed Database System(cont…)
Example:
EMPLOYEE, PROJECT and WORKS_ON tables may
be fragmented horizontally and stored with possible
replication as shown below:
4
Distributed Database System(cont…)
Advantages (cont…)
Distribution and Network transparency:
Users do not have to worry about operational details of the
network
There is Location transparency, which refers to
freedom of issuing command from any location without
affecting its working
Replication transparency:
It allows to store copies of a data at multiple sites for
better availability.
Makes the user unaware of the existence of copies
This is done to minimize access time to the required data.
Fragmentation transparency:
Allows to fragment a relation horizontally (create a subset
of tuples of a relation) or vertically (create a subset of
columns of a relation)
Makes the user unaware of the existence of fragments
5
Distributed Database System(cont…)
Advantages (cont...)
Increased reliability and availability:
Reliability refers to system life time; that is, system
is running efficiently most of the time
Availability is the probability that the system is
continuously available (usable or accessible) during
a time interval
A distributed database system has multiple nodes
(computers) and if one fails then others are
available to do the job.
6
Distributed Database System(cont…)
Other Advantages (cont…)
Improved performance:
A distributed DBMS fragments the database to keep
data closer to where it is needed most
This reduces data management overhead (access
and modification time) significantly
Easier expansion (scalability):
Refers to expansion of the system in terms of
adding more data, increasing database sizes or
adding more processors
7
Data Fragmentation, Replication and Allocation
Data Fragmentation
Split a relation into logically related and correct parts. A
relation can be fragmented in two ways:
Horizontal Fragmentation - Vertical Fragmentation
Horizontal fragmentation
It is a horizontal subset of a relation which contain those of
tuples which satisfy selection conditions.
Consider the Employee relation with selection condition (DNO
= 5). All tuples that satisfy this condition will create a subset
which will be a horizontal fragment of Employee relation.
A selection condition may be composed of several conditions
connected by AND / OR
8
Data Fragmentation, Replication and
Allocation(cont…)
Vertical fragmentation
It is a subset of a relation which is created by a subset of
columns. Thus a vertical fragment of a relation will contain
values of selected columns. There is no selection condition
used in vertical fragmentation.
Consider the Employee relation. A vertical fragment of can
be created by keeping the values of Name, Bdate, Sex, and
Address.
Because there is no condition for creating a vertical
fragment, each fragment must include the primary key
attribute of the parent relation Employee.
In this way all vertical fragments of a relation are connected.
9
Data Fragmentation, Replication and
Allocation(cont…)
Representing horizontal fragmentation
Each horizontal fragment on a relation can be specified
by a sCi (R) operation in the relational algebra
Complete horizontal fragmentation
A set of horizontal fragments whose conditions C1,
C2, …, Cn include all the tuples in R- that is, every
tuple in R satisfies
(C1 OR C2 OR … OR Cn)
Disjoint complete horizontal fragmentation: No tuple in R
satisfies (Ci AND Cj) where i ≠ j
To reconstruct R from horizontal fragments a UNION is
applied
10
Data Fragmentation, Replication and
Allocation(cont…)
Vertical fragmentation
A vertical fragment on a relation can be specified by a
Li(R) operation in the relational algebra.
Complete vertical fragmentation
A set of vertical fragments whose projection lists L1, L2,
…, Ln include all the attributes in R but share only the
primary key of R. In this case the projection lists satisfy
the following two conditions:
L1 L2 ... Ln = ATTRS (R)
Li Lj = PK(R) for any i, j, where ATTRS (R) is the set
of attributes of R and PK(R) is the primary key of R.
11
Data Fragmentation, Replication and
Allocation(cont…)
Mixed (Hybrid) fragmentation
A combination of Vertical fragmentation and Horizontal
fragmentation
This is achieved by SELECT-PROJECT operations
which is represented by Li(sCi (R))
12
Data Fragmentation, Replication and
Allocation(cont…)
Data Replication
Replication refers to the distribution of whole or part of
the data to a number of sites
Useful in improving availability of data
Improve performance of global queries since the result of
such query can be obtained from any one site
In full replication, the entire database is replicated and in
partial replication some selected part is replicated to
some of the sites
The disadvantage of full replication is that it can slow
down update operation since a single logical update must
be performed on every copy of the database to keep the
copies consistent
13
Types of Distributed Database Systems
Homogeneous
All sites of the database
system have identical Window
setup, i.e., same database Site 5 Unix
Oracle Site 1
system software. Oracle
For example, all sites run Window
Oracle or DB2, or Sybase Site 4 Communications
network
or some other but the
same database system Oracle
software.
Site 3 Site 2
The underlying operating Linux Oracle Linux Oracle
systems may be different
(can be a mixture of Linux,
Window, Unix, etc.)
14
Types of Distributed Database Systems
Heterogeneous
Federated: Each site may run different database
system but the data access is managed through a
single conceptual schema.
Multidatabase: There is no one conceptual global
schema. For data access a schema is constructed
dynamically as needed by the application software.
Object Unix Relational
Oriented Site 5 Unix
Site 1
Hierarchical
Window
Site 4 Communications
network
Network
Object DBMS
Oriented Site 3 Site 2 Relational
Linux Linux
15
Types of Distributed Database Systems
Federated Database Management Systems(FDBSs) Issues
The type of heterogeneity present in FDBSs may arise from
several sources:
Differences in data models:
Relational, Objected oriented, hierarchical, network, etc.
Differences in constraints:
Each site may have their own data accessing and
processing constraints.
Differences in query language:
Even with the data model, the language and their version
may vary. SQL has multiple versions. some may use
SQL-89, and other may use SQL-92, SQL3 and so on.
16
Query Processing in Distributed Databases
Issues
Cost of transferring data (files and results) over the
network.
This cost is usually high. So, some optimization is
necessary.
Example: Suppose we have the Employee relation at site 1
and Department relation at Site 2
Employee at site 1. 10,000 rows. Row size = 100 bytes.
6
This means, table size = 10 bytes.
Department at Site 2. 100 rows. Row size = 35 bytes.
This means, table size = 3,500 bytes.
17
Query Processing in Distributed Databases (cont…)
Issues (cont…)
Query Q : For each employee, retrieve employee name and
department name Where the employee works.
Q: Fname,Lname,Dname (EmployeeDno = Dnumber Department)
Employee Fname MName Lname SSN Bdate Address Sex Slary Superssn Dno
Department Dname Dnumber Mgrssn Mgrstartdate
18
Query Processing in Distributed Databases (cont…)
Result
Suppose that Employee and Department relations are
not present at site 3(see the figure shown below)
Suppose that each result tuple is 40 bytes long. The
query is submitted at site 3 and the result is sent to this
site
If every employee is related to a department, the result of
this query will have 10,000 tuples
Employee
Site 1
Site 2 Site 3
Department
19
Query Processing in Distributed Databases (cont…)
Strategies (Available options):
1. Transfer Employee and Department to site 3.
Total transfer bytes = 1,000,000 + 3500 = 1,003,500 bytes.
2. Transfer Employee to site 2, execute join at site 2 and send
the result to site 3.
Query result size = 40 * 10,000 = 400,000 bytes.
Total transfer size = 1,000,000 + 400,000 = 1,400,000 bytes.
3. Transfer Department relation to site 1, execute the join at
site 1, and send the result to site 3
Total bytes transferred = 3500 + 400,000 = 403,500 bytes.
Optimization criteria: minimizing data transfer.
Preferred strategy: strategy 3.
20
Query Processing in Distributed Databases (cont…)
Consider the query
Q’: For each department, retrieve the department
name and the name of the department manager
The query is submitted at site 3
Relational Algebra expression:
Fname,Lname,Dname (Department Mgrssn = SSNEmployee)
Assuming that every department has a manager,
the result of this query will have 100 tuples
21
Query Processing in Distributed Databases (cont…)
Execution strategies:
1. Transfer Employee and Department to the result site and
perform the join at site 3.
Total bytes transferred = 1,000,000 + 3500 = 1,003,500 bytes
2. Transfer Employee to site 2, execute join at site 2 and
send the result to site 3.
Query result size = 40 * 100 = 4000 bytes.
Total transfer size = 1,000,000 +4000 = 1,004,000 bytes.
3. Transfer Department relation to site 1, execute join at site
1 and send the result to site 3.
Total transfer size = 3500 + 4000 = 7500 bytes.
Preferred strategy: Choose strategy 3.
22
Query Processing in Distributed
Databases (cont…)
Now suppose the result site is 2.
Possible strategies :
1. Transfer Employee relation to site 2, execute the query and
present the result to the user at site 2
Total transfer size = 1,000,000 bytes for both queries Q and Q’.
2. Transfer Department relation to site 1, execute join at site 1
and send the result back to site 2
Total transfer size for Q:
3500 +400,000 = 403,500 bytes
Total transfer size for Q’:
3500 +4000 = 7500 bytes
23
Query Processing in Distributed Databases
using semijoin
Semijoin:
The idea behind semijoin operation is to reduce the number of tuples
in a relation before transferring it to another site.
Example: using the queries Q or Q’ discussed in the previous slides :
1. Project the join attributes of Department at site 2, and transfer them to
site 1.
Assume size of Dnumber=4 bytes and size of Mgrssn=9 bytes
Assume size of fname and lname is 15 bytes each
For Q, 4 * 100 = 400 bytes are transferred and for Q’, 9 * 100 = 900
bytes are transferred
2. Join the transferred file with the Employee relation at site 1, and
transfer the required attributes from the resulting file to site 2.
For Q, 34 * 10,000 = 340,000 bytes are transferred and
For Q’, 39 * 100 = 3900 bytes are transferred
3. Execute the query by joining the transferred file with Department and
present the result to the user at site 2.
Using this strategy, we transfer 340,400 bytes for Q and 4800 bytes for Q’
24
Concurrency Control and Recovery
Distributed Databases encounter a number of
concurrency control and recovery problems which
are not present in centralized databases.
Some of these problems are listed below:
Dealing with multiple copies of data items
Failure of individual sites
Communication link failure
Distributed commit
Distributed deadlock
25
Concurrency Control and Recovery (cont…)
Details
Dealing with multiple copies of data items:
The concurrency control must maintain global
consistency
Likewise, the recovery mechanism must recover all
copies and maintain consistency after recovery
Failure of individual sites:
Database availability must not be affected due to
the failure of one or two sites and the recovery
scheme must recover them before they are
available for use
26
Concurrency Control and Recovery (cont…)
(Details….)
Communication link failure:
This failure may create network partition which would
affect database availability even though all database sites
may be running.
Distributed commit:
Problems can arise with transactions that is accessing
databases stored on multiple sites if some sites fail during
the commit process . The 2 phase commit is used to deal
with this problem
Distributed deadlock:
Since transactions are processed at multiple sites, two or
more sites may get involved in deadlock. This must be
resolved in a distributed manner.
27
Concurrency Control and Recovery (cont…)
Distributed Concurrency control based on distinguished
copy of a data item
A. Primary site technique: A single site is assigned as
a primary site which serves as a coordinator for
transaction management.
Primary site
Site 5
Site 1
Site 4 Communications neteork
Site 3 Site 2
28
Concurrency Control and Recovery
Transaction management:
Concurrency control and commit are managed by
this site
All locks are kept at that site and all requests for
locking or unlocking are sent there
In two phase locking, this site manages locking
and releasing of data items
If all transactions follow two-phase policy at all
sites, then serializability is guaranteed
29
Concurrency Control and Recovery (cont…)
Advantages:
It is an extension to the centralized two phase locking
and hence simple to Implement and manage
Data items are locked only at one site but they can be
accessed at any site at which they reside
Disadvantages:
All transaction management activities go to primary
site which is likely to overload the site.
If the primary site fails, the entire system is
inaccessible
Primary site with backup site
To aid recovery, a backup site is designated which
behaves as a shadow of primary site.
In case of primary site failure, backup site can act as
primary site.
Slows down system performance for granting of locks
30
Concurrency Control and Recovery (cont…)
B. Primary Copy Technique:
In this approach, Distinguished copies of different data
items are stored at different sites
Load of lock coordination is distributed among
the various sites
To lock a data item, just the primary copy of the
data item is locked
Advantages:
Since primary copies are distributed at various sites, a
single site is not overloaded with locking and
unlocking requests
Failure of one site affects only transactions that are
accessing locks on items whose primary copy reside
on that site, while other transactions are not affected
31
Concurrency Control and Recovery
Recovery from a coordinator failure
In both approaches, a coordinator site or copy may become
unavailable. This will require the selection of a new
coordinator.
Primary site approach with no backup site:
Aborts and restarts all active transactions at all sites. Elects
a new coordinator and initiates transaction processing.
Primary site approach with backup site:
Suspends all active transactions, designates the backup
site as the primary site and identifies a new back up site.
Primary site receives all transaction management
information to resume processing.
Primary and backup sites fail or no backup site:
Use election process to select a new coordinator site.
For better understanding, of the election process,
Refer the text book (Fundamentals of DB
systems.Elmasri and Navathe 6th edition, page 911)
32
Concurrency Control and Recovery
Concurrency control based on voting:
In a voting method, method, there is no distinguished copy;
Rather, a lock request is sent to all sites that includes a
copy of the data item.
Each copy maintains its own lock and can grant or deny the
request for it.
If a transaction that requests a lock is granted that lock by a
majority of the copies, it holds the lock and informs all
copies that it has been granted the lock
To avoid unacceptably long wait, a time-out period is
defined.
If the requesting transaction does not get any vote
information, the transaction is aborted.
33
Client-Server Database Architecture
It consists of clients running client software, a set of
servers which provide all database functionalities and a
reliable communication infrastructure.
Server 1 Client 1
Client 2
Server 2 Client 3
Server n Client n
34
Client-Server Database Architecture
Server: is responsible for local data management at a
site, much like centralized DBMS software
Client: is responsible for most of the distribution function;
it accesses data distribution information from the DBMS
catalog and processes all requests that require access to
more than one site
The communication software manages communication
among clients and servers
35
Client-Server Database Architecture
The processing of a SQL queries goes as follows:
Client parses a user query and decomposes it into
a number of independent sub-queries.
Each server processes its query and sends the
result to the client.
The client combines the results of sub queries and
produces the final result.
36