Parallel database systems

Presented by
Basanti Bhattacharyya
Assistant Professor ( IT Department)of
St. Thomas’ College of Engineering & Technology
1
 Parallel Databases
• A parallel DBMS is a DBMS that runs across multiple processors and
is designed to execute operations in parallel, whenever possible.
The parallel DBMS links several smaller machines to achieve the
same throughput as expected from a single large machine.
• Advantages of Parallel Databases
• Increased Speed and Efficiency: That is why the application of
parallel databases allows executing several queries at the same time
and consequently, decreases the total amount of time needed for
processing them.
• Improved Resource Utilization: Most of them take advantage of
the multi-processor architecture, thus using as many CPU cores as
possible.
• High Throughput: They perform a large number of computations in
less time more than the sequential methods hence suitable for large-
scale uses.

2
 Parallel Databases

• Disadvantages of Parallel Databases

• Complexity in Maintenance: Indeed, employment of
parallel databases might be slightly problematic as
synchronization and some other components of the
processes are worthy of special attention.
• Higher Costs: While parallel databases may be defined
with robust physical databases which require intensive
hardware the databases may be more costly in setting up
and even in operation.

3
 Difference Between Parallel and Distributed Databases
Parallel Database Distributed Database

In distributed databases, the sites are loosely

In parallel databases, processes are tightly coupled and
coupled and share no physical components i.e.,
constitutes a single database system i.e., the parallel
distributed database is our geographically
database is a centralized database and data reside in a single
departed, and data are distributed at several
location
locations.

In parallel databases, query processing and transaction is In distributed databases, query processing and
complicated. transaction is more complicated.

In distributed databases, a local and global

In parallel databases, it's not applicable. transaction can be transformed into distributed
database systems

In distributed databases, each site preserve a local

In parallel databases, the data is partitioned among various
database system for faster processing due to the
disks so that it can be retrieved faster.
slow interconnection between sites

In parallel databases, there are 3 types of architecture: Distributed databases are generally a kind of
shared memory, shared disk, and shared shared-nothing. shared-nothing architecture

In distributed databases, query Optimization

In parallel databases, query optimization is more
techniques may be different at different sites and
complicated.
are easy to maintain

In distributed databases, data is replicated at any

In parallel databases, data is generally not copied. number of sites to improve the performance of
systems 4
 Characteristics of Parallel Database

Characteristics of Tightly Coupled Systems:

•Shared Resources: Processors share common memory,
disks,
• and I/O components.
•Single System Image: They appear and function as a single
database system to the user, managed by one OS.
•High Performance: High data rates and fast communication
between processors.
•High Coordination: Strong interdependence means
components
• must be active simultaneously.
•Examples: Symmetric multiprocessors (SMP) and
Massively Parallel Processors (MPP

5
 Design of Parallel Databases
• Parallel architectures in distributed systems involve using multiple
interconnected processing elements (nodes, CPUs, or machines) to
perform tasks simultaneously, significantly enhancing performance,
throughput, and scalability.
• In Parallel Databases, mainly there are three architectural designs for
parallel DBMS. They are as follows:

• Shared Memory Architecture

• Shared Disk Architecture
• Shared Nothing Architecture

6
 Shared Memory Architecture
In Shared Memory Architecture, there are multiple CPUs that are attached to an
interconnection network. They are able to share a single or global main memory
and common disk arrays. It is to be noted that, In this architecture, a single copy
of a multi-threaded operating system and multithreaded DBMS can support these
multiple CPUs. Also, the shared memory is a solid coupled architecture in which
multiple CPUs share their memory. It is also known as Symmetric
multiprocessing (SMP). This architecture has a very wide range which starts
from personal workstations that support a few microprocessors in parallel via
RISC.

Shared Memory Architecture 7

 Shared Memory Architecture (cont.…)

Advantages :
• It has high-speed data access for a limited number of
processors.
• The communication is efficient.

Disadvantages :
• It cannot use beyond 80 or 100 CPUs in parallel.
• The bus or the interconnection network gets block due to
the increment of the large number of CPUs.

8
 Shared Disk Architectures
In Shared Disk Architecture, various CPUs are attached to an
interconnection network. In this, each CPU has its own memory and all
of them have access to the same disk. Also, note that here the memory is
not shared among CPUs therefore each node has its own copy of the
operating system and DBMS. Shared disk architecture is a loosely
coupled architecture optimized for applications that are inherently
centralized. They are also known as clusters.

Shared Disk Architecture 9

 Shared Disk Architectures (cont.…)

• Advantages :
• The interconnection network is no longer a bottleneck
each CPU has its own memory.
• Load-balancing is easier in shared disk architecture.
• There is better fault tolerance.
•
Disadvantages :
• If the number of CPUs increases, the problems of
interference and memory contentions also increase.
• There's also exists a scalability problem

10
 Shared Nothing Architecture
• Shared Nothing Architecture is multiple processor architecture in
which each processor has its own memory and disk storage. In this,
multiple CPUs are attached to an interconnection network through a
node. Also, note that no two CPUs can access the same disk area. In
this architecture, no sharing of memory or disk resources is done. It is
also known as Massively parallel processing (MPP).

Shared Nothing Architecture 11

 Shared Nothing Architecture (cont.…)
• Advantages :
• It has better scalability as no sharing of resources is done
• Multiple CPUs can be added
•
Disadvantages:
• The cost of communications is higher as it involves
sending of data and software interaction at both ends
• The cost of non-local disk access is higher than the cost
of shared disk architectures

12
 Hierarchical Architecture

This architecture is a combination of shared disk, shared

memory and shared nothing architectures. This architecture
is scalable due to availability of more memory and many
processor, But is costly to other architecture.

13
 Parallel query processing
• Different queries/transactions can be run in parallel with
each other.
• Interquery parallelism
• Concurrency control takes care of conflicts in case of
updates
• More on parallel transaction processing .
• Focus in this topics is on read-only queries
• Individual relational operations (e.g., sort, join,
aggregation) can be executed in parallel
• Data can be partitioned and each processor can work
independently on its own partition.
• Queries are expressed in high level language (SQL,
translated to relational algebra
14
 Parallel Processing of Relational Operations

• Our discussion of parallel algorithms assumes:

• read-only queries
• shared-nothing architecture
• N nodes, N1, ..., Nn
--Each assumed to have disks and processors.
• Initial focus on parallelization to a shared-nothing node
--Parallel processing within a shared memory/shared disk node
discussed later.
• Shared-nothing architectures can be efficiently simulated on
shared-memory and shared-disk systems.
• Algorithms for shared-nothing systems can thus be run on
shared-memory and shared-disk systems.
• However, some optimizations may be possible.
15
THANKS

16