Distributed Systems – Architecture Models

pribblingchoppedElectronics - Devices

Nov 15, 2013 (3 years and 9 months ago)

89 views

D
ISTRIBUTED

S
YSTEMS



A
RCHITECTURE

M
ODELS


Janani C Krishnamani

CSC 8320

Fall
2011

O
UTLINE


Introduction


Architecture models


System architectures


Communication Network architectures


Examples


Future Ideas


I
NTRODUCTION

W
HAT

ARE

DISTRIBUTED

SYSTEMS
?




Collection of independent computers that appear
to the user as one single entity


Examples


Systems
in a LAN, WAN, MAN


World Wide Web


Torrent networks


Clusters


Cloud
servers



W
HY

INTERCONNECT

SYSTEMS
? [1]


To pass a message from one system to another


Share common
resources


B
eneficial
to use multiple low end processors
than one high end processor


Fault tolerance through redundancy



C
LASSIFICATION

C
LASSIFICATION

[2]


Based on system architecture


Client


Server model


Peer


Peer model


Based on communication network


Bus


Switched


Based on level of coupling


Tightly coupled (parallel)


Loosely coupled (distributed)

B
ASED

ON

S
YSTEM

A
RCHITECTURE

:
C
LIENT

S
ERVER

M
ODEL


One server acts as the principal control agent


Several nodes report to the server


Advantages:


Better control and security


Concentration of functions in high performance
servers


Disadvantages:


Not robust due to lack of redundancy


Performance suffers if the number of clients
increases


Examples:


Workstation


Server Model


Processor Pool Model

B
ASED

ON

S
YSTEM

A
RCHITECTURE

:
W
ORKSTATION



S
ERVER

M
ODEL



Many
workstations connect to the network.


Workstations provide local processing and an
interface to the network
.


Consist of one or more client workstations
connected to one or more server workstations
using a communication network.


Mainly used for resource sharing



B
ASED

ON

S
YSTEM

A
RCHITECTURE

:
P
ROCESSOR

P
OOL

M
ODEL


All the processing is moved to a pool of processors


No concept of a home machine


Workstations are dummy terminals with good
graphical display


Terminals have minimum intelligence like
remote booting, remote mounting of file systems,
virtual terminal handling and packet assembly
and disassembly services

B
ASED

ON

S
YSTEM

A
RCHITECTURE

:

P
EER



P
EER

M
ODEL


All the nodes in the network have equal privilege


No special routing server required. The nodes
themselves take care of routing the data


Peers


both suppliers and consumers


As the number of nodes increases, the bandwidth
increases


Reduces single point
failures


It is less secure


Usually used for sharing of resources


files,
audio
-

visual media

B
ASED

ON

C
OMMUNICATION

N
ETWORK

:
B
US



B
ASED

I
NTERCONNECTION


Usually used in Point


Point connection
systems


System connected by computer bus


The access to communication media is time
shared


IEEE 802 LAN Standard


Ethernet, Token Bus, Token Ring, Fiber Distributed
data Interface (FDDI), Fiber Queue
Dual
Buses
(FQDB)


B
ASED

ON

C
OMMUNICATION

N
ETWORK

:
S
WITCH



B
ASED

I
NTERCONNECTION


Usually used in multi point connection systems


Systems connected by network switches


The access to communication media is both time
and space shared


but cost equally higher than
bus


Private switches


Crossbar, multistage switch


Public switch systems


Integrated Services
Digital Network (ISDN), Asynchronous Transfer
Mode (ATM


Switches in public networks also perform routing


B
ASED

ON

L
EVEL

OF

C
OUPLING

:
L
OOSELY

C
OUPLED

S
YSTEMS


Each processor has its own memory


R
uns its own OS


Do not communicate frequently


Can be placed in geographically separate regions


Also called distributed
computing


B
ASED

ON

L
EVEL

OF

C
OUPLING

:

T
IGHTLY

C
OUPLED

S
YSTEMS


Processors
share a common memory


Run on a single OS


Communicate very frequently


Usually connected through high speed connection
methods


Also called parallel computing


L
EVEL

OF

C
OUPLING

Microprocessors

Clusters

LAN

Global Internet

Small

Fast

Large

Slow

More Tightly Coupled

More Loosely Coupled

Closer in physical proximity

Farther in physical proximity

G
RID

C
OMPUTING

[3]


Complete nodes connected by a conventional
network interface like Ethernet.


Not tightly coupled. Can be
geographically
diverse.


The communication is not high
speed
-

works
fine
in environments that require less
communication.


A single task is adequately parallelized and each
part is given to a different node that computes it
independently
-

so
no concurrency issues
involved.


C
LOUD

C
OMPUTING

W
HAT

IS

C
LOUD

C
OMPUTING

[4]




P
arallel
and distributed
system consisting
of a
collection of inter
-
connected and
virtualized
computers


Presented to the user as a single computer


Services provided:


Software as service


Platform as service


Infrastructure as service


Advantages


Speed of Operation


Power saving


Abstraction


Virtualization


E
XAMPLES

F
OLDING
@H
OME

[6]


Grid computing project
designed to perform
computationally intensive simulations of protein
folding and other molecular
dynamics


The
most powerful distributed computing cluster
in the
world


Guinness 2007


Primary contributors


thousands of everyday
personal computers


Processes run in the background utilizing
cpu

idle time


In late March 2011
Folding@home

briefly peaked
above the 7 native
petaFLOP

barrier
.

W
INDOWS

A
ZURE

P
LATFORM

[7]


On demand platform used to build, host and scale
web applications through windows datacenter.


Platform as a service


Build, host and maintain web applications
running on Microsoft Datacenter


OS


Windows Azure


Services


Live Services


SQL Azure


AppFabric


SharePoint Services


Dynamic CRM Services





F
UTURE


F
UTURE

OF

D
ISTRIBUTED

C
OMPUTING



An expanded peer
-

peer network with every
personal computer interconnected with every
other.


Utilization of CPU cycles and hard disks during
idle time


Improvements in security and performance that
void the need for a centralized control
server and
a data center


R
EFERENCES


[1]
http
://
en.wikipedia.org/wiki/Distributed_computi
ng#Architectures


[2]
http://
en.wikipedia.org/wiki/Grid_computing


[3]
http://
en.wikipedia.org/wiki/Cloud_computing


[4]
Ghosh
,
Anup
;
Arce
, Ivan; , "Guest Editors'
Introduction: In Cloud Computing We Trust
-

But
Should We?,"
Security & Privacy, IEEE

, vol.8,
no.6, pp.14
-
16, Nov.
-
Dec. 2010

doi
:
10.1109/MSP.2010.177


[5]
http://
en.wikipedia.org/wiki/Folding@Home


[6]
http://en.wikipedia.org/wiki/Blue_Gene
/


[7]
http://
en.wikipedia.org/wiki/Windows_Azure




T
HANK

Y
OU
!