1 Introduction

Distributed Database Systems
Lecture slides as adapted and customized from © 2020, M.T. Özsu & P. Valduriez 1
Outline
 Introduction
 Distributed and Parallel Database Design
 Distributed Data Control
 Distributed Query Processing
 Distributed Transaction Processing
 Data Replication
 Database Integration – Multidatabase Systems
 Parallel Database Systems
Outline
 Introduction
 What is a distributed DBMS
 History
 Distributed DBMS promises/benefits
 Design issues/challenges/problems
 Distributed DBMS architecture
Distributed Computing
 A number of autonomous processing elements (not

necessarily homogeneous) that are interconnected by a
computer network and that cooperate in performing their
assigned tasks.
 What is being distributed?
 Processing logic
 Function
 Data
 Control
Current Distribution – Geographically
Distributed Data Centers
What is a Distributed Database System?
A distributed database is a collection of multiple, logically

interrelated databases distributed over a computer network
A distributed database management system (Distributed

DBMS) is the software that manages the DDB and provides
an access mechanism that makes this distribution
transparent to the users
What is not a DDBS?
 A timesharing computer system

 A loosely or tightly coupled multiprocessor system
 A database system which resides at one of the nodes of
a network of computers - this is a centralized database
on a network node
Distributed DBMS Environment
Implicit Assumptions
 Data stored at a number of sites → each site logically

consists of a single processor
 Processors at different sites are interconnected by a
computer network → not a multiprocessor system
 Parallel database systems
 Distributed database is a database, not a collection of
files → data logically related as exhibited in the users’
access patterns
 Relational data model
 Distributed DBMS is a full-fledged DBMS
 Not remote file system, not a TP system
Important Point
Logically integrated
but
Physically distributed
Outline
 Introduction
 History
History – File Systems
History – Database Management
History – Early Distribution
Peer-to-Peer (P2P)
History – Client/Server
History – Data Integration
History – Cloud Computing
On-demand, reliable services provided over the Internet in

a cost-efficient manner
 Cost savings: no need to maintain dedicated compute
power
 Elasticity: better adaptivity to changing workload
Data Delivery Alternatives
 Delivery modes
 Pull-only
 Push-only
 Hybrid
 Frequency
 Periodic
 Conditional
 Ad-hoc or irregular
 Communication Methods
 Unicast
 One-to-many
 Note: not all combinations make sense
Outline
 Introduction
 History
Distributed DBMS promises/benefits
 Transparent management of distributed, fragmented, and

replicated data
 Improved reliability/availability through distributed

transactions
 Improved performance
 Easier and more economical system expansion
Lecture slides as adapted and customized from © 2020, M.T. Özsu & P. Valduriez
Transparency
 Transparency is the separation of the higher-level

semantics of a system from the lower level
implementation issues.
 Fundamental issue is to provide data independence
in the distributed environment
 Network (distribution) transparency
 Replication transparency
 Fragmentation transparency
 horizontal fragmentation: selection
 vertical fragmentation: projection
 hybrid
Lecture slides as adapted and customized from © 2020, M.T. Özsu & P. Valduriez
Example
Transparent Access
Tokyo
SELECT ENAME,SAL
Boston Paris
FROM EMP,ASG,PAY
WHERE DUR > 12 Paris projects
Paris employees
AND EMP.ENO = ASG.ENO Communication Paris assignments
Network Boston employees
AND PAY.TITLE =
EMP.TITLE Boston projects
Boston employees
Boston assignments
Montreal
New
Montreal projects
York Paris projects
Boston projects New York projects
New York employees with budget > 200000
New York projects Montreal employees
New York assignments Montreal assignments
Distributed Database - User View
Distributed Database
Distributed DBMS - Reality
User
Query
User
DBMS
Application
Software
DBMS
Software
DBMS Communication
Software Subsystem
User
DBMS User Application
Software Query
DBMS
Software
User
Query
Types of Transparency
 Data independence
 Network transparency (or distribution transparency)
 Location transparency
 Fragmentation transparency
 Fragmentation transparency
 Replication transparency
Reliability Through Transactions
 Replicated components and data should make distributed

DBMS more reliable.
 Distributed transactions provide
 Concurrency transparency
 Failure atomicity
• Distributed transaction support requires implementation of

 Distributed concurrency control protocols
 Commit protocols
 Data replication
 Great for read-intensive workloads, problematic for updates
 Replication protocols
Potentially Improved Performance
 Proximity of data to its points of use

 Requires some support for fragmentation and replication
 Parallelism in execution
 Inter-query parallelism
 Intra-query parallelism
Scalability
 Issue is database scaling and workload scaling
 Adding processing and storage power
 Scale-out: add more servers

 Scale-up: increase the capacity of one server → has limits
Geographic Distribution
 Distributed databases support geographic distribution of

data, allowing organizations to store data closer to users
or locations where it is needed.
 This reduces latency and improves responsiveness for

users accessing data from different geographic regions.
Issue is database scaling and workload scaling
Flexibility and Adaptability
 Distributed databases offer greater flexibility and

adaptability to changing business requirements and
environments.
 Organizations can easily add or remove nodes, adjust

data distribution strategies, and scale resources as
needed to meet evolving demands.
Data Security and Privacy
 Distributed databases incorporate robust security

mechanisms to protect data confidentiality, integrity, and
availability.
 Encryption, access controls, authentication, and auditing

features help ensure compliance with data privacy
regulations and mitigate security risks.
Cost efficiency
 Distributed databases can offer cost efficiencies

compared to centralized databases, particularly in terms
of hardware and infrastructure costs.
 By leveraging commodity hardware and cloud computing

resources, organizations can achieve cost-effective
scalability and resource utilization.
Resilience to Network Failures
 Distributed databases are designed to withstand network

failures and disruptions.
 Replication, data synchronization, and distributed

consensus algorithms ensure data consistency and
integrity even in the face of network partitions or
communication failures.
Outline
 Introduction
 History
Distributed DBMS
Issues/challenges/problems
 Distributed database design
 How to distribute the database
 Replicated & non-replicated database distribution
 Distributed query processing
 Convert user transactions to data manipulation instructions
 Optimization problem
 min{cost = data transmission + local processing}
 General formulation is NP-hard
Distributed DBMS
 Distributed concurrency control
 Synchronization of concurrent accesses
 Consistency and isolation of transactions' effects
 Deadlock management
 Reliability
 How to make the system resilient to failures: consider when a
system in inoperable or inaccessible due to network failure for
instance
 How to ensure database is up to date after recover?
 Consistency, atomicity and durability
Distributed DBMS
 Distributed directory management
 A directory may be global to the entire DDBS or local to each
site; it can be centralized at one site or distributed over several
sites; there can be a single copy or multiple copies
 Contains information (such as descriptions and locations) about
data items in database
 How to distribute the database
 Replicated & non-replicated database distribution
Distributed DBMS
 Distributed deadlock management
 Consider the competition among users for access to a set of
resources (data, in this case); this can result in a deadlock if the
synchronization mechanism is based on locking.
 How to handle deadlock in DDBSs: prevention, avoidance, and
detection/recovery.
Distributed DBMS
 Replication
 Mutual consistency
 Freshness of copies
 Eager vs lazy
 Centralized vs distributed
 Parallel DBMS
 Objectives: high scalability and performance
 Not geo-distributed
 Cluster computing
Related Issues
 Alternative distribution approaches

 Modern P2P
 World Wide Web (WWW or Web)
 Big data processing
 4V: volume, variety, velocity, veracity
 MapReduce & Spark
 Stream data
 Graph analytics
 NoSQL
 NewSQL
 Polystores
Outline
 Introduction
 History
ANSI/SPARC Architecture
 The ANSI/SPARC
architecture has three
views of data:
 the external view,
which is that of the
end user, who might
be a programmer;
 the internal view, that
of the system or
machine: physical
definition and
organization of data.
 the conceptual view,
that of the enterprise
ANSI/SPARC Architecture
 The ANSI/SPARC
architecture has three
views of data:
 the external view,
which is that of the
end user, who might
be a programmer;
 the internal view, that
of the system or
machine: physical
definition and
organization of data.
 the conceptual view,
that of the enterprise
DBMS Implementation Alternatives
Architectural Models for Distributed DBMSs
DBMS Implementation Alternatives
Architectural Models for Distributed DBMSs: Dimensions of the Problem
 Distribution
 Whether the components of the system are located on the same machine or not
 Heterogeneity
 Various levels (hardware, communications, operating system)
 DBMS important one
 data model, query language, transaction management algorithms
 Autonomy
 Refers to distribution of control
 Various versions
 Design autonomy: Ability of a component DBMS to decide on issues related to its own
design.
 Communication autonomy: Ability of a component DBMS to decide whether and how
to communicate with other DBMSs.
 Execution autonomy: Ability of a component DBMS to execute local operations in any
manner it wants to.
Client/Server Architecture
Advantages of Client-Server
Architectures
 More efficient division of labor
 Horizontal and vertical scaling of resources
 Better price/performance on client machines
 Ability to use familiar tools on client machines
 Client access to remote data (via standards)
 Full DBMS functionality provided to client workstations
 Overall better system price/performance
Database Server
Distributed Database Servers
Peer-to-Peer Component Architecture
MDBS Components & Execution
A multidatabase system, also known as a federated database system,
is a type of database management system (DBMS) that integrates
multiple autonomous databases into a single logical database.
Mediator/Wrapper Architecture
Cloud Computing
On-demand, reliable services provided over the Internet in

a cost-efficient manner
 IaaS – Infrastructure-as-a-Service
 PaaS – Platform-as-a-Service
 SaaS – Software-as-a-Service
 DaaS – Database-as-a-Service
Simplified Cloud Architecture
Distributed computing Individual exercise
a. Define the concept of distributed database management
systems (DDBMS) and explain its significance in
modern information systems.
b. Discuss the key differences between a centralized
DBMS and a distributed DBMS.
c. a. Define a distributed database management system
(DDBMS) and explain how it differs from a traditional
centralized database management system.
d. Discuss the benefits and challenges associated with
implementing a distributed DBMS in organizations.
e. Identify and discuss the key design issues and
challenges involved in building a distributed database
management system.
f. Using books and the Internet resources, explain how
design issues such as data distribution, concurrency
control, data replication, query optimization, and
security are addressed in distributed DBMS
architectures.
g. Describe the architecture of a distributed database
management system, including its components, layers,
and functionalities. Compare it with the multidatabase
systems integration
h. Discuss the role of distributed data storage, query
processing, concurrency control mechanisms,
transaction management, and data replication in
distributed DBMS architectures.
i. Using books and the Internet resources, expound on the
following characteristics of the distributed database
systems. NB: characteristics define the inherent
qualities or attributes of distributed database systems
data distribution, autonomy, heterogeneity,

transparency, concurrency control, data replication,
scalability, fault tolerance, consistency and consensus,
and security.

1 Introduction

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 Introduction

Uploaded by

Copyright:

Available Formats

Distributed Database Systems

 A number of autonomous processing elements (not

A distributed database is a collection of multiple, logically

A distributed database management system (Distributed

 A timesharing computer system

 Data stored at a number of sites → each site logically

On-demand, reliable services provided over the Internet in

 Transparent management of distributed, fragmented, and

 Improved reliability/availability through distributed

 Easier and more economical system expansion

 Transparency is the separation of the higher-level

 Replicated components and data should make distributed

• Distributed transaction support requires implementation of

 Proximity of data to its points of use

 Issue is database scaling and workload scaling

 Adding processing and storage power

 Scale-out: add more servers

 Distributed databases support geographic distribution of

 This reduces latency and improves responsiveness for

 Distributed databases offer greater flexibility and

 Organizations can easily add or remove nodes, adjust

 Distributed databases incorporate robust security

 Encryption, access controls, authentication, and auditing

 Distributed databases can offer cost efficiencies

 By leveraging commodity hardware and cloud computing

 Distributed databases are designed to withstand network

 Replication, data synchronization, and distributed

 Alternative distribution approaches

On-demand, reliable services provided over the Internet in

data distribution, autonomy, heterogeneity,

You might also like