SlideShare a Scribd company logo
1 of 35
Interconnection Mechanisms
Performance Models
Connecting Processors and Memories
• Shared Buses
• Interconnection Networks
• Static Networks
• Dynamic Networks
slide 2
P P P P
M M M
Interconnection Network
M
M M M
P P P P
M M M
Interconnection Network
M
M M M
Global Interconnection Network
M M M
Shared Bus
slide 3
each processor sees this picture:
processing
bus access
timentransactiobustimeprocessing
timentransactiobus
nutilizatiobus
+
== ρ
prob of a processor using the bus = ρ
prob of a processor not using the bus = 1 – ρ
prob of none of the n processors using the bus = (1 – ρ)n
prob of at least one processor using the bus = 1 – (1 – ρ)n
achieved BW on a relative scale = 1 – (1 – ρ)n
required BW = n ρ available BW = 1
Effect of re-submitted requests
slide 4
A W
ρ (1-PA )1- ρ + ρPA 1-PA
PA
( ) ( )
( ) ( ) ( )
( )
( )ρ
ρ
ρ
ρ
ρ
ρ
ρ
ρ
ρρ
ρ
ρρρρ
ρ
ρ
ρρρ
−+
=⇒=−−==
−+
=
−+
−+
−+
=
+==
−=
−+
=
−+
=
1
also11
11
1
1
raterequestactual
1
11
a
a
A
n
a
AA
A
A
A
wA
AW
A
A
AA
A
A
aPanBW
PP
P
P
P
qqa
qq
P
P
PP
P
q
prob = qA prob = qW
Shared Bus :BWper proc
-0.100
0.000
0.100
0.200
0.300
0.400
0.500
0.600
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
BW required (req probability)
BWachieved
n = 2
n = 3
n = 4
n = 2
n = 3
n = 4
Shared Bus :utilization
-0.200
0.000
0.200
0.400
0.600
0.800
1.000
1.200
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
req probability
utilization
n = 2
n = 3
n = 4
n = 2
n = 3
n = 4
Waiting time
slide 7
[ ]
bus
a
a
bus
A
A
A
A
Abus
i
A
i
Abus
A
i
A
i
busw
A
i
A
th
bus
TT
P
P
P
P
PTPiPT
PPTiT
PP
)(ii
Ti
ρ
ρρ −
=
−
=
−−
−
××=−×××=
×−××==
×−=
+
×=
∑
∑
∞
=
∞
=
1
)1(1
1
)1(
)1(timewaitingofvalueExpected
)1(thisofyprobabilit
attempt1onacceptedandtimesrejectedisrequestif
timewaiting
2
1
1
Switched Networks
BUS
• Shared media
• Lower Cost
• Lower throughput
• Scalability poor
Switched Network
• Switched paths
• Higher cost
• Higher throughput
• Scalability better
slide 8
Interconnection Networks• Topology : who is connected to whom
• Direct / Indirect : where is switching done
• Static / Dynamic : when is switching done
• Circuit switching / packet switching : how are
connections established
• Store & forward / worm hole routing : how is the path
determined
• Centralized / distributed : how is switching controlled
• Synchronous/asynchronous : mode of operation
slide 9
Direct and Indirect Networks
slide 10
P
M
P
M
S
P
M
S
S
M
P
S
M
P
P
M
P
M
P
M
SWITCH
DIRECT
INDIRECT
node node
node node
link
link
link link
node
node
node
node
link
link
link
link
Static and Dynamic Networks
• Static Networks
• fixed point to point connections
• usually direct
• each node pair may not have a direct connection
• routing through nodes
• Dynamic Networks
• connections established as per need
• usually indirect
• path can be established between any pair of nodes
• routing through switches
slide 11
Static Network Topologies
slide 12
Linear
Star
2D-Mesh
Tree
Non-uniform connectivity
Static Networks Topologies-
contd.
slide 13
Ring
Fully Connected
Torus
Uniform connectivity
Illiac IV Mesh Network
slide 14
0 1 2
3 4 5
6 7 8
0
1
2
3
45
6
7
8
neighbors of node r :
(r ± 1) mod 9 and
(r ± 3) mod 9 Chordal Ring
Fat Tree Network
slide 15
Dynamic Networks
slide 16
k × k
cross -bar
switch
building block for multi-stage
dynamic networks
2 × 2
switch
straight exchange upper
broadcast
lower
broadcast
simplest
cross-bar
Baseline Network
slide 17
000
001
010
011
100
101
110
111
000
001
010
011
100
101
110
111
blocking can occur
Benes Network
slide 18
non-blocking
Switching Mechanism
• Circuit Switching (connection oriented
communication)
• A circuit is established between the source and the
destination
• Packet Switching (connectionless communication)
• Information is divided into packets and each packet is
sent independently from node to node
slide 19
Routing in Networks
slide 20
node
incoming
message
outgoing
message
header payload/data
store & forward
routing
worm hole
routing
time
BW
H
BW
l






+×=
BW
l
BW
H
nlatency
BW
l
BW
H
nlatency +





×=
Routing in presence of congestion
• Worm hole routing
• When message header is blocked, many links get
blocked with the message
• Solution: cut-through routing
• When message header is blocked, tail is allowed to
move, compressing the message into a single node
slide 21
Routing Options
• Deterministic routing: always same path followed
• Adaptive routing: best path selected to minimize
congestion
• Source based routing: message specifies path to
destination
• Destination based routing: message specifies only
destination address
slide 22
Some Performance
Parameters
slide 23
time
sender
receiver
time of flight
overhead
overhead
Tx time=bytes/BW
Tx time=bytes/BW
transport latency
total latency
Other Parameters
• Throughput ≈ Bandwidth (no credit for header)
• Bisection bandwidth = BW across a bisection
• Node degree
• Network Diameter
• Cost
• Fault Tolerance
slide 24
Multidimensional Grid/Mesh
Size
=k × k × …. × k (n times)
= k n
Diameter
= (k-1) × n without end around
connections
= k × n /2 with end around
connections
slide 25
k-ary n-cube
for (Binary) Hypercube : k = 2
Grid/Mesh Performance - 1
slide 26
cycleainreqmessageofprobis
dimensiononealong
hopsofno.av.is
dimensionsofnumberis
ratearrivalMessage
r
k
n
knr
d
d=
=
λ
kd
Grid/Mesh Performance - 2
n
p
Tkr
T
n
sd
s
2
linkaalong
requestofyProbabilit
2
OccupancyServer
2
rateService
ρ
µ
λ
ρ
µ
=
==
=
slide 27
Grid/Mesh Performance - 3
slide 28
k-ary n-cube
sw
w
T
pp
T
D
T
)1(2)1(2
)(1
modelqueueopen1//Muse
nodeaattimewaiting
B
ρ
ρ
ρ
ρρ
λ −
−
=





−
−
=
=
Switch Performance
slide 29
k × m
cross -bar
switch
m
m
m
m
m
mm
E(i)i
rrCq(i)ki
T
r
i
i
ii
iki
i
k













 −
−=
−−
=
×
=
=
−==
=
=
−
1
1
)1(
portsoutputofnum
portoutputspecificaincludingpatternsaddressoffraction
requestsofoutacceptedrequestsofno.expected
)1(portsonrequestsussimultaneoofprob
timeservicesamerequirespacket)(or
mesageeachthatassumedisitHere
cycleserviceoneduringport
inputanatrequestofprobLet
Switch Performance – contd.
slide 30
kk
k
i
ik
i
i
k
k
i
iki
i
k
k
i
iki
i
k
ik
i
iki
i
k
k
i
iki
i
k
i
k
i
m
r
mmrr
m
m
mm
rr
m
m
CmrrCm
rrCm
m
m
rrCm
rrCm
m
m
iqiE






−−=





−+
−
−=
−




 −
−−=
−













 −
−−=
−













 −
−=
=
∑∑
∑∑
∑
∑
=
−
=
−
=
−
=
−
=
−
=
1)1(
1
)1(
1
)1(
)1(
1
)1(
)1(
1
1
)()(scale)relative(onBWExpected
00
00
0
0
Switch Performance – contd.
slide 31
waiting.ofbecausedelayscomputealsoandsubmission-re
todueraterequestrevisedcomputetoneedWe
conflicts.toduesubmission-rerequestofeffectconsidernowWe
requestsofacceptanceofprob
)1that(assumingaswellasthanlessisthis
1conflicts)portoutputof(becauseBWExpected
conflicts)portoutputnowerethere(ifBWExpected
bandwidthRequested
kr
BW
P
rr km
m
r
mm
m
r k
A
k
==
<






−−=
=
=
Effect of re-submitted requests
slide 32
( )
linkofBW
1
timewaiting
'
'
1
1
'
)andstatesgraph withMarkov(using
'raterequestactual
lH
timecycleT
T
P
P
kr
BW
P
m
r
mmBW
rPr
r
r
qq
qqrr
A
A
A
k
A
wA
wA
+
==
−
=
=





−−=
−+
=⇒
+==
Effect of buffering
There are two possibilities
• Buffering before switching (k buffers, one at each
input port)
• Buffering after switching (m buffers, one at each
output port)
slide 33
Switch with input buffers
Rate of messages at input and output of each
queue is same in steady state - r per cycle
Service time includes delays due to conflicts,
calculated as earlier. This has an
exponential distribution – recall the analysis for a
shared bus.
M/M/1 open queue model can be used to calculate
queuing delay. Details are omitted.
T
P
P
A
A−1
slide 34
Switch with output buffers
Here we assume that all the messages destined for same
output are queued in the same buffer, in some order. That
is no rejections and no re-submissions.
For each queue,
Messages arriving per service cycle = ρ =
Prob of a request coming from one of
the k sources = p =
Apply MB/D/1 model for finding queuing delay Tw
m
kr
m
r
slide 35
T
p
Tw
)1(2 ρ
ρ
−
−
=

More Related Content

What's hot

Data communication lab manual
Data communication lab manualData communication lab manual
Data communication lab manual
Nafe Singh Yadav
 
Flynns classification
Flynns classificationFlynns classification
Flynns classification
Yasir Khan
 
Point To Point Protocol
Point To Point ProtocolPoint To Point Protocol
Point To Point Protocol
Phan Vuong
 
Routing protocols-network-layer
Routing protocols-network-layerRouting protocols-network-layer
Routing protocols-network-layer
Nitesh Singh
 

What's hot (20)

Data communication lab manual
Data communication lab manualData communication lab manual
Data communication lab manual
 
System interconnect architecture
System interconnect architectureSystem interconnect architecture
System interconnect architecture
 
Communication costs in parallel machines
Communication costs in parallel machinesCommunication costs in parallel machines
Communication costs in parallel machines
 
Flynns classification
Flynns classificationFlynns classification
Flynns classification
 
system interconnect architectures in ACA
system interconnect architectures in ACAsystem interconnect architectures in ACA
system interconnect architectures in ACA
 
Point To Point Protocol
Point To Point ProtocolPoint To Point Protocol
Point To Point Protocol
 
Routing protocols in ad hoc network
Routing protocols in ad hoc networkRouting protocols in ad hoc network
Routing protocols in ad hoc network
 
Issues in routing protocol
Issues in routing protocolIssues in routing protocol
Issues in routing protocol
 
Aca2 01 new
Aca2 01 newAca2 01 new
Aca2 01 new
 
Multi Processors And Multi Computers
 Multi Processors And Multi Computers Multi Processors And Multi Computers
Multi Processors And Multi Computers
 
Computer network switches & their structures
Computer network switches & their structuresComputer network switches & their structures
Computer network switches & their structures
 
COA (Unit_4.pptx)
COA (Unit_4.pptx)COA (Unit_4.pptx)
COA (Unit_4.pptx)
 
Pipeline hazards in computer Architecture ppt
Pipeline hazards in computer Architecture pptPipeline hazards in computer Architecture ppt
Pipeline hazards in computer Architecture ppt
 
Distributed design alternatives
Distributed design alternativesDistributed design alternatives
Distributed design alternatives
 
Routing protocols-network-layer
Routing protocols-network-layerRouting protocols-network-layer
Routing protocols-network-layer
 
Shortest path algorithm
Shortest  path algorithmShortest  path algorithm
Shortest path algorithm
 
NUMA overview
NUMA overviewNUMA overview
NUMA overview
 
All-Reduce and Prefix-Sum Operations
All-Reduce and Prefix-Sum Operations All-Reduce and Prefix-Sum Operations
All-Reduce and Prefix-Sum Operations
 
Introduction to Distributed System
Introduction to Distributed SystemIntroduction to Distributed System
Introduction to Distributed System
 
Message passing in Distributed Computing Systems
Message passing in Distributed Computing SystemsMessage passing in Distributed Computing Systems
Message passing in Distributed Computing Systems
 

Viewers also liked

Networking
NetworkingNetworking
Networking
ryanette
 
Interconnecting devices
Interconnecting devices Interconnecting devices
Interconnecting devices
har139
 
Parallel computing chapter 2
Parallel computing chapter 2Parallel computing chapter 2
Parallel computing chapter 2
Md. Mahedi Mahfuj
 
Bengali optical character recognition system
Bengali optical character recognition systemBengali optical character recognition system
Bengali optical character recognition system
Md. Mahedi Mahfuj
 

Viewers also liked (8)

Interconnection Network
Interconnection NetworkInterconnection Network
Interconnection Network
 
Networking
NetworkingNetworking
Networking
 
Interconnection 101
Interconnection 101Interconnection 101
Interconnection 101
 
Interconnections: SPICESS
Interconnections:  SPICESSInterconnections:  SPICESS
Interconnections: SPICESS
 
Interconnecting devices
Interconnecting devices Interconnecting devices
Interconnecting devices
 
Parallel computing chapter 2
Parallel computing chapter 2Parallel computing chapter 2
Parallel computing chapter 2
 
Telecommunications Policies Standards and Regulations Notes
Telecommunications Policies Standards and Regulations NotesTelecommunications Policies Standards and Regulations Notes
Telecommunications Policies Standards and Regulations Notes
 
Bengali optical character recognition system
Bengali optical character recognition systemBengali optical character recognition system
Bengali optical character recognition system
 

Similar to Interconnection mechanisms

4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
mrcopyxerox
 
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOCSOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
SnehaLatha68
 

Similar to Interconnection mechanisms (20)

Computer Organozation
Computer OrganozationComputer Organozation
Computer Organozation
 
Chapter10 switching
Chapter10 switchingChapter10 switching
Chapter10 switching
 
Network layer
Network layerNetwork layer
Network layer
 
Part 4 : reliable transport and sharing resources
Part 4 : reliable transport and sharing resourcesPart 4 : reliable transport and sharing resources
Part 4 : reliable transport and sharing resources
 
Data Link Layer of OSI Model responsibilities
Data Link Layer of OSI Model responsibilitiesData Link Layer of OSI Model responsibilities
Data Link Layer of OSI Model responsibilities
 
Data_Link_Layer.ppt
Data_Link_Layer.pptData_Link_Layer.ppt
Data_Link_Layer.ppt
 
Switching concepts Data communication and networks
Switching concepts Data communication and networksSwitching concepts Data communication and networks
Switching concepts Data communication and networks
 
Performance Analysis of Creido Enhanced Chord Overlay Protocol for Wireless S...
Performance Analysis of Creido Enhanced Chord Overlay Protocol for Wireless S...Performance Analysis of Creido Enhanced Chord Overlay Protocol for Wireless S...
Performance Analysis of Creido Enhanced Chord Overlay Protocol for Wireless S...
 
Lec 04 - Gate-level Minimization
Lec 04 - Gate-level MinimizationLec 04 - Gate-level Minimization
Lec 04 - Gate-level Minimization
 
Lecture14
Lecture14Lecture14
Lecture14
 
SCOR: Constraint Programming-based Northbound Interface for SDN
SCOR: Constraint Programming-based Northbound Interface for SDNSCOR: Constraint Programming-based Northbound Interface for SDN
SCOR: Constraint Programming-based Northbound Interface for SDN
 
4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
4af46e43-4dc7-4b54-ba8b-3a2594bb5269 j.pdf
 
MERITA PROJECT:METHODOLOGY TO EVALUATE RADIATED EMISSION FROM HIGH DENSITY MU...
MERITA PROJECT:METHODOLOGY TO EVALUATE RADIATED EMISSION FROM HIGH DENSITY MU...MERITA PROJECT:METHODOLOGY TO EVALUATE RADIATED EMISSION FROM HIGH DENSITY MU...
MERITA PROJECT:METHODOLOGY TO EVALUATE RADIATED EMISSION FROM HIGH DENSITY MU...
 
module10-rip (1).ppt
module10-rip (1).pptmodule10-rip (1).ppt
module10-rip (1).ppt
 
Is Multipath Routing Really a Panacea?
Is Multipath Routing Really a Panacea?Is Multipath Routing Really a Panacea?
Is Multipath Routing Really a Panacea?
 
Network layer
Network layerNetwork layer
Network layer
 
Underwater communications using magnetic induction
Underwater communications using magnetic inductionUnderwater communications using magnetic induction
Underwater communications using magnetic induction
 
1 chayes
1 chayes1 chayes
1 chayes
 
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOCSOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
SOC-CH5.pptSOC Processors Used in SOCSOC Processors Used in SOC
 
Consistent Resource Scheduling and QoS management
Consistent Resource Scheduling and QoS managementConsistent Resource Scheduling and QoS management
Consistent Resource Scheduling and QoS management
 

More from Syed Zaid Irshad

Basic Concept of Information Technology
Basic Concept of Information TechnologyBasic Concept of Information Technology
Basic Concept of Information Technology
Syed Zaid Irshad
 
Introduction to ICS 1st Year Book
Introduction to ICS 1st Year BookIntroduction to ICS 1st Year Book
Introduction to ICS 1st Year Book
Syed Zaid Irshad
 

More from Syed Zaid Irshad (20)

Operating System.pdf
Operating System.pdfOperating System.pdf
Operating System.pdf
 
DBMS_Lab_Manual_&_Solution
DBMS_Lab_Manual_&_SolutionDBMS_Lab_Manual_&_Solution
DBMS_Lab_Manual_&_Solution
 
Data Structure and Algorithms.pptx
Data Structure and Algorithms.pptxData Structure and Algorithms.pptx
Data Structure and Algorithms.pptx
 
Design and Analysis of Algorithms.pptx
Design and Analysis of Algorithms.pptxDesign and Analysis of Algorithms.pptx
Design and Analysis of Algorithms.pptx
 
Professional Issues in Computing
Professional Issues in ComputingProfessional Issues in Computing
Professional Issues in Computing
 
Reduce course notes class xi
Reduce course notes class xiReduce course notes class xi
Reduce course notes class xi
 
Reduce course notes class xii
Reduce course notes class xiiReduce course notes class xii
Reduce course notes class xii
 
Introduction to Database
Introduction to DatabaseIntroduction to Database
Introduction to Database
 
C Language
C LanguageC Language
C Language
 
Flowchart
FlowchartFlowchart
Flowchart
 
Algorithm Pseudo
Algorithm PseudoAlgorithm Pseudo
Algorithm Pseudo
 
Computer Programming
Computer ProgrammingComputer Programming
Computer Programming
 
ICS 2nd Year Book Introduction
ICS 2nd Year Book IntroductionICS 2nd Year Book Introduction
ICS 2nd Year Book Introduction
 
Security, Copyright and the Law
Security, Copyright and the LawSecurity, Copyright and the Law
Security, Copyright and the Law
 
Computer Architecture
Computer ArchitectureComputer Architecture
Computer Architecture
 
Data Communication
Data CommunicationData Communication
Data Communication
 
Information Networks
Information NetworksInformation Networks
Information Networks
 
Basic Concept of Information Technology
Basic Concept of Information TechnologyBasic Concept of Information Technology
Basic Concept of Information Technology
 
Introduction to ICS 1st Year Book
Introduction to ICS 1st Year BookIntroduction to ICS 1st Year Book
Introduction to ICS 1st Year Book
 
Using the set operators
Using the set operatorsUsing the set operators
Using the set operators
 

Recently uploaded

Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
dollysharma2066
 
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoorTop Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
dharasingh5698
 
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
amitlee9823
 

Recently uploaded (20)

Unleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapUnleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leap
 
Introduction to Serverless with AWS Lambda
Introduction to Serverless with AWS LambdaIntroduction to Serverless with AWS Lambda
Introduction to Serverless with AWS Lambda
 
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
VIP Model Call Girls Kothrud ( Pune ) Call ON 8005736733 Starting From 5K to ...
 
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
(INDIRA) Call Girl Meerut Call Now 8617697112 Meerut Escorts 24x7
 
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
FULL ENJOY Call Girls In Mahipalpur Delhi Contact Us 8377877756
 
FEA Based Level 3 Assessment of Deformed Tanks with Fluid Induced Loads
FEA Based Level 3 Assessment of Deformed Tanks with Fluid Induced LoadsFEA Based Level 3 Assessment of Deformed Tanks with Fluid Induced Loads
FEA Based Level 3 Assessment of Deformed Tanks with Fluid Induced Loads
 
22-prompt engineering noted slide shown.pdf
22-prompt engineering noted slide shown.pdf22-prompt engineering noted slide shown.pdf
22-prompt engineering noted slide shown.pdf
 
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoorTop Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
Top Rated Call Girls In chittoor 📱 {7001035870} VIP Escorts chittoor
 
A Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna MunicipalityA Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna Municipality
 
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdf
 
COST-EFFETIVE and Energy Efficient BUILDINGS ptx
COST-EFFETIVE  and Energy Efficient BUILDINGS ptxCOST-EFFETIVE  and Energy Efficient BUILDINGS ptx
COST-EFFETIVE and Energy Efficient BUILDINGS ptx
 
Thermal Engineering Unit - I & II . ppt
Thermal Engineering  Unit - I & II . pptThermal Engineering  Unit - I & II . ppt
Thermal Engineering Unit - I & II . ppt
 
Employee leave management system project.
Employee leave management system project.Employee leave management system project.
Employee leave management system project.
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
 
2016EF22_0 solar project report rooftop projects
2016EF22_0 solar project report rooftop projects2016EF22_0 solar project report rooftop projects
2016EF22_0 solar project report rooftop projects
 
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
 
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
 

Interconnection mechanisms

  • 2. Connecting Processors and Memories • Shared Buses • Interconnection Networks • Static Networks • Dynamic Networks slide 2 P P P P M M M Interconnection Network M M M M P P P P M M M Interconnection Network M M M M Global Interconnection Network M M M
  • 3. Shared Bus slide 3 each processor sees this picture: processing bus access timentransactiobustimeprocessing timentransactiobus nutilizatiobus + == ρ prob of a processor using the bus = ρ prob of a processor not using the bus = 1 – ρ prob of none of the n processors using the bus = (1 – ρ)n prob of at least one processor using the bus = 1 – (1 – ρ)n achieved BW on a relative scale = 1 – (1 – ρ)n required BW = n ρ available BW = 1
  • 4. Effect of re-submitted requests slide 4 A W ρ (1-PA )1- ρ + ρPA 1-PA PA ( ) ( ) ( ) ( ) ( ) ( ) ( )ρ ρ ρ ρ ρ ρ ρ ρ ρρ ρ ρρρρ ρ ρ ρρρ −+ =⇒=−−== −+ = −+ −+ −+ = +== −= −+ = −+ = 1 also11 11 1 1 raterequestactual 1 11 a a A n a AA A A A wA AW A A AA A A aPanBW PP P P P qqa qq P P PP P q prob = qA prob = qW
  • 5. Shared Bus :BWper proc -0.100 0.000 0.100 0.200 0.300 0.400 0.500 0.600 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BW required (req probability) BWachieved n = 2 n = 3 n = 4 n = 2 n = 3 n = 4
  • 6. Shared Bus :utilization -0.200 0.000 0.200 0.400 0.600 0.800 1.000 1.200 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 req probability utilization n = 2 n = 3 n = 4 n = 2 n = 3 n = 4
  • 7. Waiting time slide 7 [ ] bus a a bus A A A A Abus i A i Abus A i A i busw A i A th bus TT P P P P PTPiPT PPTiT PP )(ii Ti ρ ρρ − = − = −− − ××=−×××= ×−××== ×−= + ×= ∑ ∑ ∞ = ∞ = 1 )1(1 1 )1( )1(timewaitingofvalueExpected )1(thisofyprobabilit attempt1onacceptedandtimesrejectedisrequestif timewaiting 2 1 1
  • 8. Switched Networks BUS • Shared media • Lower Cost • Lower throughput • Scalability poor Switched Network • Switched paths • Higher cost • Higher throughput • Scalability better slide 8
  • 9. Interconnection Networks• Topology : who is connected to whom • Direct / Indirect : where is switching done • Static / Dynamic : when is switching done • Circuit switching / packet switching : how are connections established • Store & forward / worm hole routing : how is the path determined • Centralized / distributed : how is switching controlled • Synchronous/asynchronous : mode of operation slide 9
  • 10. Direct and Indirect Networks slide 10 P M P M S P M S S M P S M P P M P M P M SWITCH DIRECT INDIRECT node node node node link link link link node node node node link link link link
  • 11. Static and Dynamic Networks • Static Networks • fixed point to point connections • usually direct • each node pair may not have a direct connection • routing through nodes • Dynamic Networks • connections established as per need • usually indirect • path can be established between any pair of nodes • routing through switches slide 11
  • 12. Static Network Topologies slide 12 Linear Star 2D-Mesh Tree Non-uniform connectivity
  • 13. Static Networks Topologies- contd. slide 13 Ring Fully Connected Torus Uniform connectivity
  • 14. Illiac IV Mesh Network slide 14 0 1 2 3 4 5 6 7 8 0 1 2 3 45 6 7 8 neighbors of node r : (r ± 1) mod 9 and (r ± 3) mod 9 Chordal Ring
  • 16. Dynamic Networks slide 16 k × k cross -bar switch building block for multi-stage dynamic networks 2 × 2 switch straight exchange upper broadcast lower broadcast simplest cross-bar
  • 19. Switching Mechanism • Circuit Switching (connection oriented communication) • A circuit is established between the source and the destination • Packet Switching (connectionless communication) • Information is divided into packets and each packet is sent independently from node to node slide 19
  • 20. Routing in Networks slide 20 node incoming message outgoing message header payload/data store & forward routing worm hole routing time BW H BW l       +×= BW l BW H nlatency BW l BW H nlatency +      ×=
  • 21. Routing in presence of congestion • Worm hole routing • When message header is blocked, many links get blocked with the message • Solution: cut-through routing • When message header is blocked, tail is allowed to move, compressing the message into a single node slide 21
  • 22. Routing Options • Deterministic routing: always same path followed • Adaptive routing: best path selected to minimize congestion • Source based routing: message specifies path to destination • Destination based routing: message specifies only destination address slide 22
  • 23. Some Performance Parameters slide 23 time sender receiver time of flight overhead overhead Tx time=bytes/BW Tx time=bytes/BW transport latency total latency
  • 24. Other Parameters • Throughput ≈ Bandwidth (no credit for header) • Bisection bandwidth = BW across a bisection • Node degree • Network Diameter • Cost • Fault Tolerance slide 24
  • 25. Multidimensional Grid/Mesh Size =k × k × …. × k (n times) = k n Diameter = (k-1) × n without end around connections = k × n /2 with end around connections slide 25 k-ary n-cube for (Binary) Hypercube : k = 2
  • 26. Grid/Mesh Performance - 1 slide 26 cycleainreqmessageofprobis dimensiononealong hopsofno.av.is dimensionsofnumberis ratearrivalMessage r k n knr d d= = λ kd
  • 27. Grid/Mesh Performance - 2 n p Tkr T n sd s 2 linkaalong requestofyProbabilit 2 OccupancyServer 2 rateService ρ µ λ ρ µ = == = slide 27
  • 28. Grid/Mesh Performance - 3 slide 28 k-ary n-cube sw w T pp T D T )1(2)1(2 )(1 modelqueueopen1//Muse nodeaattimewaiting B ρ ρ ρ ρρ λ − − =      − − = =
  • 29. Switch Performance slide 29 k × m cross -bar switch m m m m m mm E(i)i rrCq(i)ki T r i i ii iki i k               − −= −− = × = = −== = = − 1 1 )1( portsoutputofnum portoutputspecificaincludingpatternsaddressoffraction requestsofoutacceptedrequestsofno.expected )1(portsonrequestsussimultaneoofprob timeservicesamerequirespacket)(or mesageeachthatassumedisitHere cycleserviceoneduringport inputanatrequestofprobLet
  • 30. Switch Performance – contd. slide 30 kk k i ik i i k k i iki i k k i iki i k ik i iki i k k i iki i k i k i m r mmrr m m mm rr m m CmrrCm rrCm m m rrCm rrCm m m iqiE       −−=      −+ − −= −      − −−= −               − −−= −               − −= = ∑∑ ∑∑ ∑ ∑ = − = − = − = − = − = 1)1( 1 )1( 1 )1( )1( 1 )1( )1( 1 1 )()(scale)relative(onBWExpected 00 00 0 0
  • 31. Switch Performance – contd. slide 31 waiting.ofbecausedelayscomputealsoandsubmission-re todueraterequestrevisedcomputetoneedWe conflicts.toduesubmission-rerequestofeffectconsidernowWe requestsofacceptanceofprob )1that(assumingaswellasthanlessisthis 1conflicts)portoutputof(becauseBWExpected conflicts)portoutputnowerethere(ifBWExpected bandwidthRequested kr BW P rr km m r mm m r k A k == <       −−= = =
  • 32. Effect of re-submitted requests slide 32 ( ) linkofBW 1 timewaiting ' ' 1 1 ' )andstatesgraph withMarkov(using 'raterequestactual lH timecycleT T P P kr BW P m r mmBW rPr r r qq qqrr A A A k A wA wA + == − = =      −−= −+ =⇒ +==
  • 33. Effect of buffering There are two possibilities • Buffering before switching (k buffers, one at each input port) • Buffering after switching (m buffers, one at each output port) slide 33
  • 34. Switch with input buffers Rate of messages at input and output of each queue is same in steady state - r per cycle Service time includes delays due to conflicts, calculated as earlier. This has an exponential distribution – recall the analysis for a shared bus. M/M/1 open queue model can be used to calculate queuing delay. Details are omitted. T P P A A−1 slide 34
  • 35. Switch with output buffers Here we assume that all the messages destined for same output are queued in the same buffer, in some order. That is no rejections and no re-submissions. For each queue, Messages arriving per service cycle = ρ = Prob of a request coming from one of the k sources = p = Apply MB/D/1 model for finding queuing delay Tw m kr m r slide 35 T p Tw )1(2 ρ ρ − − =