Upload
owen-garrison
View
216
Download
0
Tags:
Embed Size (px)
Citation preview
11
Under the guidance ofUnder the guidance ofProf. Sridhar IyerProf. Sridhar Iyer
Student: Annanda Th. RATHStudent: Annanda Th. RATH
HSM: A Hybrid Streaming Mechanism HSM: A Hybrid Streaming Mechanism for for
Delay-Tolerant Multimedia ApplicationsDelay-Tolerant Multimedia Applications
22
Key wordsKey words
Delay tolerant ApplicationsDelay tolerant Applications Client specifies the playback timeClient specifies the playback time Service must be done according to client requirements Service must be done according to client requirements
Pure Streaming Mechanism (PSM)Pure Streaming Mechanism (PSM) There is only one streaming server at the sourceThere is only one streaming server at the source Server is responsible for serving all the requestsServer is responsible for serving all the requests
Hybrid Streaming Mechanism (HSM)Hybrid Streaming Mechanism (HSM) Streaming from strategically chosen relay node in stead Streaming from strategically chosen relay node in stead
of central (source) serverof central (source) server Data flow is divided into two parts: FTP flow from source Data flow is divided into two parts: FTP flow from source
to selected relay node and streaming flow from selected to selected relay node and streaming flow from selected relay node to clientrelay node to client
33
Problem DefinitionProblem Definition
Objective (Goal)Objective (Goal) Improving the performance of streaming Improving the performance of streaming
serviceservice Maximizing the number of serviced clientsMaximizing the number of serviced clients Improving the delivered stream rate at the clientImproving the delivered stream rate at the client
Maximizing the bandwidth utilization in the Maximizing the bandwidth utilization in the backbone networkbackbone network
Reducing the traffic in the networkReducing the traffic in the network
44
Solution OutlineSolution Outline
Proposed a Hybrid Streaming Mechanism Proposed a Hybrid Streaming Mechanism (HSM)(HSM) Advantage of HSMAdvantage of HSM
Improving the performance of streaming serviceImproving the performance of streaming service Increasing the number of serviced clientsIncreasing the number of serviced clients Maximizing the bandwidth utilizationMaximizing the bandwidth utilization Reducing the work load at central server as well as the Reducing the work load at central server as well as the
traffic in the networktraffic in the network Where HSM can be used?Where HSM can be used?
Streaming in the internet Streaming in the internet Distance Education ProgramDistance Education Program Corporate TrainingCorporate Training
55
MotivationMotivation
Why HSM and not PSM?Why HSM and not PSM? Disadvantages of PSM Disadvantages of PSM
Clients occupy entire links in the path while Clients occupy entire links in the path while streaming, for the duration of the streamstreaming, for the duration of the stream
For the duration that a client is using the links, For the duration that a client is using the links, others who are sharing the link (s) in the path, others who are sharing the link (s) in the path, are not able to use the link (s)are not able to use the link (s)
Fewer users are servicedFewer users are serviced Links can be underutilizedLinks can be underutilized
66
Motivating ExampleMotivating Example
S
9
8
2
5
1011
64
1
3
12
7
C1 C2 C3
C14
C6
C13
C5 C4
C7C8C9
C10C11C12
Central server
Relay node
Region node
Client
768 Kbps
512 Kbps 512 Kbps
384 Kbps
256 Kbps
384 Kbps 512 Kbps
512 Kbps
384 Kbps
384 Kbps
512 Kbps
512 Kbps512 Kbps256 Kbps
384 Kbps
256 Kbps
128 Kbps384 Kbps
384 Kbps256 Kbps
512 Kbps
384 Kbps256 Kbps
384 Kbps768 Kbps
768 Kbps
Clients RequestTime(Minutes)
Requirements(Delay-Tolerance, rate)
PSM HSM
C1 0 (30,256) Served Served
C14 10 (60,256) Not served Served
C6 75 (30,256) Not served Served
C9 75 (05,480) Not served Served
C12 75 (30,256) Not served Served
50 mns
75 mns
320 Kbps 240 Kbps
240Kbps
466Kbps
448 Kbps
192 Kbps
320Kbps
480Kbps
480 Kbps
480 Kbps
PSM !
HSM !
77
Motivating Example: Cont.dMotivating Example: Cont.d
Links in thepath of client
C1
Link Bandwidth(Kbps)
DeliveredStream rate atC1(Kbps)
Unused bandwidth(Kbps)
Link busy period (Minutes)
PSM HSM PSM HSM
S -1 768 320 448 0 120 50
1-2 512 320 192 0 120 75
2-4 256 320 0 0 150 150
4-10 384 320 64 64 120 120
4-C1 256 320 0 0 150 150
Details of client C1
88
AssumptionAssumption
Links have dedicated bandwidth Links have dedicated bandwidth provisioned for the given applicationprovisioned for the given application
Selected intermediate nodes have the Selected intermediate nodes have the streaming capabilitystreaming capability
Multicasting is also supported in the Multicasting is also supported in the given network topologygiven network topology
99
Functional Overview of HSMFunctional Overview of HSM
HSM’s componentsHSM’s components Delivered stream rate calculationDelivered stream rate calculation Streaming point selectionStreaming point selection Content transferring and streamingContent transferring and streaming
Client
PSM!HSM! Content is
storing
SP
1010
Delivered Stream Rate CalculationDelivered Stream Rate Calculation
Rate calculationRate calculation Find the weakest link along the path.Find the weakest link along the path. Client delay tolerance.Client delay tolerance. Streaming durationStreaming duration
Bn+1
1
Lm+1
S SP
CL1
B1
d1
n
1
dn
m
1111
Streaming Point SelectionStreaming Point Selection
Selection strategiesSelection strategies At the node with maximum outgoing linksAt the node with maximum outgoing links
If the delivered stream rate is less than the If the delivered stream rate is less than the weakest link in the path from source to weakest link in the path from source to region noderegion node
At the node below the weakest linkAt the node below the weakest link If the delivered stream rate is greater than If the delivered stream rate is greater than
the weakest link in the path from source to the weakest link in the path from source to region noderegion node
1212
Reasoning for the selection Reasoning for the selection strategiesstrategies
S
9
8
2
5
1011
64
1
3
12
7
C1 C2 C3
C14
C6
C13
C5 C4
C7C8C9
C10C11C12
Central server
Relay node
Region node
Client
768 Kbps
512 Kbps 512 Kbps
384 Kbps
256 Kbps
384 Kbps512 Kbps
512 Kbps
384 Kbps
384 Kbps
512 Kbps
512 Kbps512 Kbps256 Kbps
384 Kbps
256 Kbps
128 Kbps384 Kbps
384 Kbps256 Kbps
512 Kbps
384 Kbps256 Kbps
384 Kbps768 Kbps768 Kbps
Clients Request(Minutes)
Client’srequirement(Delay-tolerant,
rate)
Service Strategy
C2 0 (90,128) Strategy 1
C4 15 (30,128)
C11 15 (30,128)
C14 15 (60,128)
C1 0 (90,256) Strategy 2
C3 100 (30,256)
C13 110 (30,256)
224 Kbps448 Kbps
1313
Note: Streaming Point SelectionNote: Streaming Point Selection According to the two strategies, there are two According to the two strategies, there are two
possible places for streaming point: possible places for streaming point: (i) at the node, which has the maximum outgoing links and (i) at the node, which has the maximum outgoing links and (ii) at the node below the weakest link from the source to (ii) at the node below the weakest link from the source to
region noderegion node Given the random nature of the clients’ request Given the random nature of the clients’ request
and its requirements, it is hard to predict the and its requirements, it is hard to predict the delivered stream rate at the client, it may happen delivered stream rate at the client, it may happen that some time, delivered stream rate is less than that some time, delivered stream rate is less than the weakest link and some time it is greater than the weakest link and some time it is greater than the weakest linkthe weakest link
In order to cover the two cases, in HSM, we deploy In order to cover the two cases, in HSM, we deploy at least two streaming points for one region, one at at least two streaming points for one region, one at the node which has the maximum outgoing links the node which has the maximum outgoing links and other at the node below the weakest link and other at the node below the weakest link
Note that one streaming point may serve more Note that one streaming point may serve more than one regionthan one region
1414
Content transferring and Content transferring and streamingstreaming Use FTP to transfer streaming content to the Use FTP to transfer streaming content to the
selected SPselected SP Content is temporarily stored at the SP for a period Content is temporarily stored at the SP for a period
equivalent to the streaming duration (SD)equivalent to the streaming duration (SD) Time To Live of Content (TTLC) is extended if new Time To Live of Content (TTLC) is extended if new
request arrivesrequest arrives TTLC expires if there is no new request within SDTTLC expires if there is no new request within SD
1515
Time to Transfer the ContentTime to Transfer the Content
Time to transfer content using FTPTime to transfer content using FTP
Bn+1
1
Lm+1
S SP
CL1
B1
d1
n
1
dn
m
1616
Caching Memory Management at Relay Caching Memory Management at Relay Node (s)Node (s) Memory requirement at relay node (s)Memory requirement at relay node (s)
A step-by-step approach to find out the caching memory size A step-by-step approach to find out the caching memory size at the relay node (s)at the relay node (s)
Find the number of sub trees rooted at SP. Let this be NFind the number of sub trees rooted at SP. Let this be N Find the number of regions in each sub tree. Let this Find the number of regions in each sub tree. Let this
number be Rnumber be R For each sub treeFor each sub tree
Find the weakest link for region j, BminjFind the weakest link for region j, Bminj Find the max of weakest link across all regions R, Wi= Find the max of weakest link across all regions R, Wi=
Max (Bminj), j=1, ..,R and i= 1, …, NMax (Bminj), j=1, ..,R and i= 1, …, N The maximum amount of data that can flow in the sub The maximum amount of data that can flow in the sub
tree i = Wi*SD, where SD is the stream durationtree i = Wi*SD, where SD is the stream duration Let FSmax be the maximum file size across all content files Let FSmax be the maximum file size across all content files
stored at S. Since clients in the regions can specify delay stored at S. Since clients in the regions can specify delay tolerance, we must find the largest file size that need to be tolerance, we must find the largest file size that need to be cached. Thus, the cache size for a sub tree is given by: cached. Thus, the cache size for a sub tree is given by: Max(Wi * SD, FSmax)Max(Wi * SD, FSmax)
The cache size at streaming point in node g, considering all The cache size at streaming point in node g, considering all the sub trees is given by:the sub trees is given by:
1717
Time To Live of ContentTime To Live of Content
How to choose TTLC and how to extend itHow to choose TTLC and how to extend it TTLC is equal the streaming duration plus TTLC is equal the streaming duration plus
client delay toleranceclient delay tolerance It is extended if there is a new request within It is extended if there is a new request within
its life timeits life time The content is removed from the cache when The content is removed from the cache when
its TTLC expiresits TTLC expires
Extension of TTLC = Tc+ CDk-(Tc-tk)+SD
Tc : is the TTLC of the current contenttk : is the time when client k’s request arrives CDk :is the delay tolerance of client k
1818
HSM based Tool ArchitectureHSM based Tool Architecture
HSM based Tool componentsHSM based Tool components
Operation Module
Network topologyand specific link bandwidth
Object name and the region that
client belongs to
Clients’ requirements(Delay Tolerance, Minimum Rate)
Delivered Stream Rate Calculation
+ Weakest link along the path detection+ Stream Rate Calculation
Streaming Point Selection+ Weakest link along the path detection+ Node with the maximum outgoing links detection
Content Transferring and streaming
+ Calculation time to transfer the content across all the links in the path+ Setting the TTLC
Outputs Module
+ Delivered stream rate at the clients
+ Link Busy Period along the path from source to client.+ TTLC
+ Streaming Point Position (Intermediate Node number)
2
1
Top Level HSM’s Architecture
1: Delivered Stream Rate 2: Streaming Point Position
Inputs Module
1919
Simulator’s Architecture Simulator’s Architecture (Matlab (Matlab Implementation)Implementation)
Topology Generator Static Network Topology
. Static network topology. Number of nodes in the network.. Number of links in the network . Link bandwidth . Level Number
Outputs
. Number of serviced clients
. Percentage improvement of client stream rate
PSM Module (1). Client’s request pattern generator. Client’s requirements generator. PSM like operation module. Number of serviced client calculation module . Percentage of stream rate improvement calculation module
HSM module (2). Client’s request pattern generator. Client’s requirements generator. HSM like operation module. Number of serviced client calculation module . Percentage of stream rate improvement calculation module
Link Bandwidth Generator Module
. Number of nodes in the network
Topology Generator
Module
. Level Number(The deep of the network)
. Link bandwidth interval (Min - Max)
Simulator’s Architecture
2020
Performance Evaluation of HSMPerformance Evaluation of HSM Simulation Parameters Simulation Parameters
Play out duration is set to 2 hrsPlay out duration is set to 2 hrs Observation period is 4 hrsObservation period is 4 hrs Queuing and propagation delay is set to zeroQueuing and propagation delay is set to zero Arrival rate of clients’ requests is varied from 1 to 30 per minuteArrival rate of clients’ requests is varied from 1 to 30 per minute
ScenariosScenarios 100 differences topologies are used, divided into two different 100 differences topologies are used, divided into two different
classes.classes. Class 1: 50 topologies, with the bandwidth in the range (256-768 Class 1: 50 topologies, with the bandwidth in the range (256-768
Kbps)Kbps) Class 2: 50 topologies, with the bandwidth in the range (128-256 Class 2: 50 topologies, with the bandwidth in the range (128-256
Kbps)Kbps) Gnutella like network topology has been used, it consists of 510 Gnutella like network topology has been used, it consists of 510
nodes with 14 levelsnodes with 14 levels Performance evaluation parametersPerformance evaluation parameters
Number of serviced clientsNumber of serviced clients Percentage of delivered stream rate improvement for a client as Percentage of delivered stream rate improvement for a client as
compared with its minimum rate requirementcompared with its minimum rate requirement
2121
Number of serviced clients Vs. Client Number of serviced clients Vs. Client Delivered Stream Rate (Class 1)Delivered Stream Rate (Class 1)
Number of serviced clients
0
20
40
60
80
100
120
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minute
Per
cent
age
of s
ervi
ced
clie
nts
HSM PSM
Stream rate improvement
130135140145150155160165170
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minute
Per
cen
tag
e o
f st
ream
ra
te i
mp
rove
men
t
HSM PSM
2222
Impact of clients’ delay-tolerance on Impact of clients’ delay-tolerance on System Performance (Class 1)System Performance (Class 1)
Number of serviced clients
0
20
40
60
80
100
120
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minute
Perc
enta
ge o
f ser
vice
d cl
ient
s
30 mns 60 mns 90 mns 120 mns
Number of serviced clients, PSM
0
20
40
60
80
100
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minutePe
rcen
tage
of s
ervi
ced
clie
nts
30 mns 60 mns 90 mns 120 mns
HSM PSM
2323
Number of serviced clients Vs. Number of serviced clients Vs. Client Delivered Stream Rate Client Delivered Stream Rate (Class 2)(Class 2)
Serviced clients VS stream rate
0
20
40
60
80
100
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minute
Perc
en
tag
e o
f se
rvic
ed
clie
nts
11.91212.112.212.312.412.512.6
Perc
en
tag
e o
f st
ream
rat
e im
pro
vem
ent
Serviced clients, HSM Serviced clients, PSM
Stream rate, HSM Stream rate, PSM
2424
Number of serviced clients Vs. Number of serviced clients Vs. Client Delivered Stream Rate Client Delivered Stream Rate (Gnutella)(Gnutella)
Serviced clients VS stream rate
0
20
40
60
80
100
120
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Number of client requests per minute
Perc
en
tag
e o
f serv
iced
cli
en
ts
142
144
146
148
150
152
154
156
158
160
162
Perc
en
tag
e o
f str
eam
ra
te i
mp
rovem
en
t
Serviced clients, HSM Serviced clients, PSM
Stream rate, HSM Stream rate, PSM
2525
Analysis of ResultsAnalysis of Results
Performance of HSM depends on two Performance of HSM depends on two factorsfactors Network topology with specific link bandwidthsNetwork topology with specific link bandwidths Clients’ requirementsClients’ requirements
HSM performs well with class1 networkHSM performs well with class1 network Because of FTP property built in HSMBecause of FTP property built in HSM
With class 2 networkWith class 2 network HSM performs slightly better than PSMHSM performs slightly better than PSM
2626
Related WorkRelated Work
Related workRelated work Initial-latency (delay) is the main issueInitial-latency (delay) is the main issue Considers streaming service as a real Considers streaming service as a real
time applicationtime application Content replication Content replication
Object placement, proxy cachingObject placement, proxy caching Resource Sharing Resource Sharing
batching, patching, interval caching, broadcastingbatching, patching, interval caching, broadcasting
Caching location problemCaching location problem
2727
Conclusions & HSM ExtensionsConclusions & HSM Extensions
ConclusionsConclusions In general, HSM performs better than PSM for the In general, HSM performs better than PSM for the
number of serviced users.number of serviced users. HSM serves more users (on the average 40%) as HSM serves more users (on the average 40%) as
compared with PSM. However it performs slightly less compared with PSM. However it performs slightly less (on the average 5%) than PSM for the streaming rate (on the average 5%) than PSM for the streaming rate improvement. improvement.
HSM can be deployed in the Distance Education Program HSM can be deployed in the Distance Education Program or streaming service in the Internet.or streaming service in the Internet.
HSM extensionsHSM extensions Admission Control.Admission Control. Converting from Static to dynamic link bandwidth.Converting from Static to dynamic link bandwidth. Optimal Placement of the streaming point.Optimal Placement of the streaming point.
2828
ReferencesReferences ““Determining the Optimal Placement for Web Proxy Cache Servers Determining the Optimal Placement for Web Proxy Cache Servers
Considering Latency in the NetworkConsidering Latency in the Network”, Vikram Tiwari, ”, Vikram Tiwari, Venkataramanam, Srinagesh GavirneniVenkataramanam, Srinagesh Gavirneni
““Caching Location ProblemCaching Location Problem”, P.Krishnan, Danny Raz, Yuval Shavitt, ”, P.Krishnan, Danny Raz, Yuval Shavitt, IEEE/ACM Transaction on Networking, VOL.8 & NO.5, October 2000.IEEE/ACM Transaction on Networking, VOL.8 & NO.5, October 2000.
““Batch Patch Caching for Streaming MediaBatch Patch Caching for Streaming Media”, Pascal Frossard and ”, Pascal Frossard and Olivier Verscheure, IEEE Communications Letter, VOL.6, NO.4Olivier Verscheure, IEEE Communications Letter, VOL.6, NO.4
““Proxy that Transcode and Cache in Heterogeneous Web Client Proxy that Transcode and Cache in Heterogeneous Web Client EnvironmentEnvironment”, Aameek Singh, Abhishek Trivedi, Krithi Ramamrithan, ”, Aameek Singh, Abhishek Trivedi, Krithi Ramamrithan, IIT Bombay.IIT Bombay.
““Streaming approach over internet: Approaches and directionsStreaming approach over internet: Approaches and directions””, , W. Z. W. Z. Y.-Q. Z. a. J. M. P. Depeng Wu, Yiwei Tomas Hou. IEEE Transaction on Y.-Q. Z. a. J. M. P. Depeng Wu, Yiwei Tomas Hou. IEEE Transaction on circuit and system for video technology, 11(3):282{300, March 2001.circuit and system for video technology, 11(3):282{300, March 2001.
““Efficient bandwidth resource allocation for low-delay multiuser video Efficient bandwidth resource allocation for low-delay multiuser video streamingstreaming”, W. W. Guang Ming Su. IEEE Transaction for Circuits and ”, W. W. Guang Ming Su. IEEE Transaction for Circuits and Systems for Video Technology, 15(9):1124{1137, September 2005.Systems for Video Technology, 15(9):1124{1137, September 2005.
““Multipath routing for video delivery over bandwidth-limited networkMultipath routing for video delivery over bandwidth-limited network”, ”, S.-H. G. C. Victor O.K, Li Jiancong Chen. IEEE Trans, S.-H. G. C. Victor O.K, Li Jiancong Chen. IEEE Trans, 22(10):1920{1932, 2004. 22(10):1920{1932, 2004.
““Optimal chaining scheme for video-on-demand applications on Optimal chaining scheme for video-on-demand applications on collaborative networkscollaborative networks”, C.-L. C. Te-Shou Su, Shih-Yu Huang and J.-S. ”, C.-L. C. Te-Shou Su, Shih-Yu Huang and J.-S. Wang. IEEE Transactions on multimedia, 7(5):972{980, October 2005.Wang. IEEE Transactions on multimedia, 7(5):972{980, October 2005.
2929
Thanks!Thanks!
Questions?Questions?
3030
Impact of client’s delay toleranceImpact of client’s delay tolerance
Why performance of HSM increases Why performance of HSM increases when client’ delay tolerance increases?when client’ delay tolerance increases?
T1 Ts1 Tf1
C1 (T1, Ts1, Tf1)
C2 (T2, Ts2, Tf2 )
T2 T2 T2
C2’s delay tolerance
3131
Delivery Stream Rate CalculationDelivery Stream Rate Calculation
assuming the streaming assuming the streaming duration is 2 hours duration is 2 hours
256 Kbps, 30 minutes delay 256 Kbps, 30 minutes delay tolerancetolerance
(S-1-2-4-10-C1) is the path (S-1-2-4-10-C1) is the path from server to C1.from server to C1.
Min (768, 512, 256, 384, Min (768, 512, 256, 384, 256) = 256 Kbps256) = 256 Kbps
30*60*256/2*3600 + 256 = 30*60*256/2*3600 + 256 = 320 Kbps320 Kbps
S
9
8
2
5
1011
64
1
3
12
7
C1 C2 C3
C14
C6
C13
C5 C4
C7C8C9
C10C11C12
768 Kbps
512 Kbps 512 Kbps
384 Kbps
256 Kbps
384 Kbps 512 Kbps
512 Kbps
384 Kbps
384 Kbps
512 Kbps
512 Kbps512 Kbps256 Kbps
384 Kbps
256 Kbps
128 Kbps384 Kbps
384 Kbps256 Kbps
512 Kbps
384 Kbps256 Kbps
384 Kbps768 Kbps
768 Kbps
50 mns
75 mns
240Kbps
3232
Caching requirement at relay Caching requirement at relay nodesnodes Let SD= 2hours (7200 seconds) Let SD= 2hours (7200 seconds) FSmax= 2 GB.FSmax= 2 GB. Let node 2 be the chosen streaming point.Let node 2 be the chosen streaming point. We calculate the cache size at node 2 as follows:We calculate the cache size at node 2 as follows:
Number of sub trees rooted at node 2, N=3Number of sub trees rooted at node 2, N=3 For sub tree 1, Number of regions, R=1; For sub tree 1, Number of regions, R=1;
Bmin1= Min (384,512)=384Bmin1= Min (384,512)=384 Max(Bmin1)= 384.Max(Bmin1)= 384.
For sub tree 2, Number of regions, R=2; For sub tree 2, Number of regions, R=2; Bmin1= Min (256, 384) =256; Bmin1= Min (256, 384) =256; Bmin2= Min (256, 256) =256;Bmin2= Min (256, 256) =256; Max (Bmin1, Bmin2)= (256, 256)= 256.Max (Bmin1, Bmin2)= (256, 256)= 256.
For sub tree 3, Number of regions, R=1;For sub tree 3, Number of regions, R=1; Bmin1= Min (384, 512, 512)=384;Bmin1= Min (384, 512, 512)=384; Max (Bmin1) = 384 Max (Bmin1) = 384
Cache size at node 2 (CS2) = Max (384*7200, 2 GB) + Cache size at node 2 (CS2) = Max (384*7200, 2 GB) + Max (256*7200, 2 GB) + Max (384*7200, 2 GB) = 7.36 GbMax (256*7200, 2 GB) + Max (384*7200, 2 GB) = 7.36 Gb
3333
Gnutella Peer NetworkGnutella Peer Network
3434
Time To Live of ContentTime To Live of Content
ExampleExample: : Let C1 with 30-minute delay tolerance Let C1 with 30-minute delay tolerance
requesting a stream with 2 hours duration.requesting a stream with 2 hours duration. TTLC for this stream is T = 30 + 120 =150 TTLC for this stream is T = 30 + 120 =150
minutes.minutes. Let a new request for the same stream comes Let a new request for the same stream comes
from client C2 at t=90 laterfrom client C2 at t=90 later C2’s delay tolerance is 90 minutes. C2’s delay tolerance is 90 minutes. The extended value of TTLC for the stream is: The extended value of TTLC for the stream is:
150 + 90 - (150 - 90) + 120. Thus, the content 150 + 90 - (150 - 90) + 120. Thus, the content is alive till t=300 minutes. is alive till t=300 minutes.
3535
Example: Admission ControlExample: Admission Control
S
9
8
2
5
1011
64
1
3
12
7
C1 C2 C3
C14
C6
C13
C5 C4
C7C8C9
C10C11C12
768 Kbps
512 Kbps 512 Kbps
384 Kbps
256 Kbps
384 Kbps 512 Kbps
512 Kbps
384 Kbps
384 Kbps
512 Kbps
512 Kbps512 Kbps256 Kbps
384 Kbps
256 Kbps
128 Kbps384 Kbps
384 Kbps256 Kbps
512 Kbps
384 Kbps256 Kbps
384 Kbps768 Kbps
768 Kbps
240Kbps
Admission Control Window time
Determine the Determine the threshold value (threshold value (by by CSPCSP))
Find the mode value Find the mode value across the collected across the collected samplessamples
If mode is less than If mode is less than the threshold valuethe threshold value
Apply the median as Apply the median as the common stream the common stream rate to all the clients rate to all the clients
Otherwise, use modeOtherwise, use mode