Transcript Distributed Systems Architecture Presentation II

Distributed Systems Architecture
Presentation II
Presenters
Rose Kit & Turgut Tezir
Outline
General View TABLE
Centralized System Architecture
Distributed System Architecture (DSA)
Advantages of DSA
Client / Server System
– Client / Server Architecture
Three-Tier client Architecture
– Three-Tier Client Server System
– Three-Tier Internet Banking System
Distributed Object Computing
Distributed Object Architecture
Advantages of Distributed Object Architecture
Distributed Object Programming Techniques
Outline, cont.
Distributed Network Organization
Network Schematic
Protocol Layers
ISO OSI Protocol Layers
TCP/IP Network Model
How OSI relates to TCP/IP Layers
Network Schematic
Internet Protocol (IP)
Transmission Control Protocol (TCP)
User Datagram Protocol (UDP)
Communication
Message passing
Sockets
Difficult To Reuse Code
Distributed Object Architecture and TCP
Distributed Object Architecture and C/S Model
Middleware
Distributed Object Architecture Model
Disadvantages of Distributed Object Architecture
RMI Example
General View TABLE
Centralized System Architecture
Centralized systems: Data, Process and
Interface components of an information
system are central. Users interact with the
system via terminals or terminal
emulators.
– A single computer with one ore more CPUs
processes all incoming request
– Problems with cost, reliability
– Specification and implementation are defined
within a single system
Centralized System Architecture
Centralized System Architecture
Distributed System Architecture
Distributed System: Data, Process, and Interface
components of an information system are distributed to
multiple locations in a computer network. Accordingly,
the processing workload is distributed across the
network.
– Set of separate computers that are capable of autonomous
operation, link by a computer network.
– Enable individual computers (different location) to share
resources in the network
– Server implementation for the same interface located in different
servers.
Distributed System Architecture
Advantages of DSA
Companies are preferring their system decentralized and
distributed, because Distributed System allows
companies to have better customer services.
– Shareability: Allows systems to use each other’s resources
– Expandability: Permits new systems to be added as members
of the overall system
– Local Autonomy: Manage local resources
– Improved performance: Resource replication. Combined
processing power of multiple computers provides much more
processing power than a centralised system with multiple CPUs
– Improved reliability and availability: Disruption would not stop
the whole system from providing its services as resources
spread across multiple computers
– Potential cost reductions
Client / Server
Client/server is a distributed computing model in which
client applications request services from server
processes. Clients and servers typically run on different
computers interconnected by a computer network.
– Client application is a process or program that sends messages
to a server via the network. Those messages request the server
to perform a specific task, such as looking up a customer record
in a database or returning a portion of a file on the server’s hard
disk.
– Server process or program listens for client requests that are
transmitted via the network. Servers receive those requests and
perform actions such as database queries and reading files.
Client / Server System
Client / Server Architecture
Three-tier Architectures
An important design consideration for large client/server systems is
whether a client talks directly to the server, or whether an
intermediary process is introduced in-between the client and the
server. The former is a two-tier architecture, the latter is a three-tier
architecture.
In the three-tier architecture, process between Server and client
(intermediary) process is:
– separate the clients and servers.
– cache frequently accessed server data to ensure better performance
and scalability.
– Performance can be increased by having the intermediary process to
distribute client requests to several servers such that requests execute
in parallel.
– The intermediary can also act as a translation service by converting
requests and replies to and from a mainframe format, or as a security
service that grants server-access only to trusted clients.
Three-Tier Client Server System
Three-Tier Internet Banking System
Distributed Object Computing
Distributed object computing is Object-oriented modeling
and programming that is applied to the development of
client/server systems.
– Objects are pieces of software that make distributed object
computing accessible through a well defined interface.
– The interface consists of object operations and attributes that are
remote accessible.
– Client applications may connect to a remote instance of the
interface and invoke the operations on the remote object. The
remote object acts as a server.
– Client/server objects have to interact with each other even if they
are written in different programming languages and to run on
different hardware and operating system platforms.
Distributed Object Architecture
There is no distinction in a distributed
object architecture between clients and
servers
Each distributable entity is an object that
provides services to other objects and
receives services from other objects
Object communication is through a
middleware system called an object
request broker (software bus)
Distributed Object Architecture
Advantages of Distributed Object
Architecture
It allows the system designer to delay decisions
on where and how services should be provided
It is a very open system architecture that allows
new resources to be added to it as required
The system is flexible and scaleable
It is possible to reconfigure the system
dynamically with objects migrating across the
network as required
Distributed Network Organization
Network can be organized over a WAN
WAN consists of a series of LANs
connected together through routers
Individual nodes (machines) in a LAN are
connected to each other through a hub
Each machine on a network can function
as the server for a particular set of
services
Network Schematic
Computer 2
LAN
Computer 1
Computer 3
Hub
Router
WAN
Router
Hub
LAN
Computer 4
Computer 5
Protocol Layers
Protocols are the standard rules for
communications between processes
Network communications requires
protocols to cover high-level application
communication all the way down to wire
communication
Complexity is handled by encapsulation in
protocol layers
TCP/IP Network Model
TCP/IP layers combine layers of ISO/OSI
TCP/IP Network Model
Networking is implemented as a series of
layered protocols starting at the lowest level (the
electronic signals on the wire and the physical
cable) up through the higher levels.
TCP/IP Layers
–
–
–
–
Layer 1:
Layer 2:
Layer 3:
Layer 4:
Link (devices and drivers)
Network (IP)
Transport (TCP)
Application (Provide network services)
How OSI relates to TCP/IP Layers
Network Schematic
Computer 2
Computer 1
Computer 3
Hub
Router
WAN
Router
Hub
Computer 4
Computer 5
Network Schematic
Computer 2
Computer 1
Computer 3
Hub
Router
WAN
Router
Hub
Computer 4
Computer 5
Network Schematic
Computer 2
Computer 1
Computer 3
Hub
Router
WAN
Router
Hub
Computer 4
Computer 5
Internet Protocol (IP)
IP is the lowest level of network
communication that handles packet
communication between machines
IP is an unreliable protocol where a
message sent is not guaranteed to reach
the intended recipient
Higher level protocols use IP as their basic
communication scheme adding reliability
and other useful features
Transmission Control Protocol
(TCP)
TCP is a higher level protocol built on top of IP.
The vast majority of network traffic take place
using the TCP protocol.
TCP is connection-based – a connection
between two nodes is established up front and
then messages may be sent in both directions.
The TCP address is a combination of an IP
address that identifies a machine and a port on
that machine. Using ports, it’s possible to
participate in multiple independent network
communications, even to communicate between
different processes on the same machine.
User Datagram Protocol (UDP)
UDP is also extends IP with the addition of ports
(like TCP).
UDP is connectionless and unreliable (like IP).
UDP is less complex and performs better than
TCP.
UDP is often used in implementing client/server
applications in distributed systems built over
local area networks where the physical
connection is over a shorter distance and there
is less chance of losing packets.
Communication
Sharing data is easy when communicating
between two processes that are in the
same address space.
Messages provide for communication
without shared data.
Message - a piece of information that is
passed from one process to another.
One process or the other owns the data at
any point, never both at the same time.
Message passing
Sockets
Sockets is a method for communication between
a client program and a server program in a
network
Communicating with sockets is low-level where
both the client and server are responsible for
ensuring that the TCP or UDP protocol is
followed correctly
Both client and server must correctly code
header information about the message, the
message size, the IP address and port number,
etc.
Difficult to reuse code
User Interface
Client 1
New 1…
Business Logic
Adapter 1
Data
Management
Adapter n
Server
New n
Distributed Object Architecture and
TCP
To allow for programming and distributing
objects across distributed networks, TCP
is used with a middleware system
– At the TCP level, all the connections between
objects on networked computers appear
identical because the details of routing across
hubs, routers, etc. are all handled at lower
levels
– Object communication is through a
middleware system
Distributed Object Architecture and
C/S Model
Client applications are represented by
objects
Server applications are represented by
objects
Back-end services like database can also
be represented as objects
The role between client and server is not
necessarily fixed in a network
Middleware
A software layer that sits between applications
and the network operating system to hide the
different underlying platforms
Provides more distribution transparency
Instead of communicating through operations on
low level sockets and using the interfaces of the
local file system, middleware offers services for
different machines to pass each other messages
at a higher abstraction level.
Java RMI, CORBA, and DCOM
Distributed Object Architecture
Model
Distributed Object Architecture
Model
Distributed Object Architecture
Model
Distributed Object Architecture
Model
Distributed Object Architecture
Model
Disadvantages of Distributed
Object Architecture
Multiple protocols may be involved
Client, server, and back-end services may have
to understand, interpret, and translate between
multiple protocols
Data structures on the client, server, and backend may conflict with each other and with
standard protocols
Client, server, and back-end may each be
written in a different programming language
Distributed Object Programming
Techniques
Java supports both RMI and CORBA
– Remote Method Invocation allows objects in a
network to remotely invoke methods of other objects
in the same network
– Objects can be transported across the network using
CORBA’s standard protocol, IIOP
– All CORBA objects are filtered through an Object
Request Broker (ORB) which allows objects to be
translated from one language to another
RMI Example
x = remoteObj.MethodA(param);
Remote Object
Stub
Remote Reference Layer
Transport Layer
RMI calls a method of the remote object
RMI Example
x = remoteObj.MethodA(param);
Stub
Remote Object
Receive
r
Remote Reference Layer
Transport Layer
Stub is generated which then sends
marshalled data to receiver
RMI Example
x = remoteObj.MethodA(param);
Stub
Remote Object
Receive
r
Remote Reference Layer
Transport Layer
Receiver unmarshals, or decodes, parameters,
locates object and calls the method specified by
the stub
RMI Example
x = remoteObj.MethodA(param);
Stub
Remote Object
Receive
r
Remote Reference Layer
Transport Layer
Receiver retrieves and marshals return value and
sends back the encoded info to the stub