Deployment of Videoconferencing in Data Networks
description
Transcript of Deployment of Videoconferencing in Data Networks
1
Deployment of Videoconferencing in Deployment of Videoconferencing in Data NetworksData Networks
Dr. Khaled SalahDr. Khaled Salah
[email protected]@kfupm.edu.sa
K. Salah 2
OutlineOutline
Guidelines can be used for any real-time Guidelines can be used for any real-time network servicenetwork service
Commercial tools to deploy VC and their Commercial tools to deploy VC and their limitationslimitations
Simulation of VC with fixed and empirical Simulation of VC with fixed and empirical packet sizespacket sizes
Delay and BW requirements for VCDelay and BW requirements for VC OPNET configurationsOPNET configurations Discussion of simulation resultsDiscussion of simulation results
K. Salah 3
K. Salah 4
K. Salah 5
DrawbacksDrawbacks
A separate add-on productA separate add-on product Usually for “naïve” engineers
Does not provide answer to how many VoIP calls can be Does not provide answer to how many VoIP calls can be supported?supported? The number of customers are a give priori
What about a new service such videoconferencing or What about a new service such videoconferencing or any other new network service?any other new network service? Shall wait for OPNET to develop a product?
Must know how to capitalize and utilize more already Must know how to capitalize and utilize more already available product available product How OPNET can be leveraged to support any network service?
K. Salah 6
OPNET and VideoconferencingOPNET and Videoconferencing
To date, OPNET does not have built-in To date, OPNET does not have built-in features or product to support deployment features or product to support deployment of videoconferencing. of videoconferencing.
We will show how to model and configure We will show how to model and configure OPNET for such a purpose. OPNET for such a purpose.
Two types of video traffic are consideredTwo types of video traffic are considered Fixed and empirical video packet sizes Empirical video packet sizes are collected from
well-known Internet traffic traces.
K. Salah 7
BackgroundBackground
The deployment of videoconferencing over IP network in both The deployment of videoconferencing over IP network in both industry and academia has been increasing rapidly. industry and academia has been increasing rapidly.
Desktop videoconferencing applications range from internal Desktop videoconferencing applications range from internal company communications, educating and training remote company communications, educating and training remote employees, to telecommuting. employees, to telecommuting.
It can eliminate certain travel requirements, thereby cutting costs. It can eliminate certain travel requirements, thereby cutting costs. Desktop videoconferencing takes advantage of a key workplace tool Desktop videoconferencing takes advantage of a key workplace tool that is the PC. that is the PC.
In the past few years, an H.323 standard was introduced by the ITU, In the past few years, an H.323 standard was introduced by the ITU, and thus paved the way to the fast growth and deployment of and thus paved the way to the fast growth and deployment of videoconferencing. videoconferencing. H.323 is a full suite of protocols developed by ITU to define how real-
time multimedia communications, such as videoconferencing, can be exchanged over packet-switched networks
K. Salah 8
It is very advantageous and cost effective to deploy It is very advantageous and cost effective to deploy desktop videoconferencing over their existing IP desktop videoconferencing over their existing IP networks. networks. It is easier to run, manage, and maintain.
However, one has to keep in mind that IP networks are However, one has to keep in mind that IP networks are best-effort networks that were designed for non-real time best-effort networks that were designed for non-real time applications. applications. On the other hand, videoconferencing requires timely packet
delivery with low latency, jitter, packet loss, and sufficient bandwidth.
To achieve this goal, an efficient deployment of videoconferencing must ensure these real-time traffic requirements can be guaranteed over new or existing IP networks.
K. Salah 9
H.320 – Videoconferencing over ISDN: Business Quality
H.321 - Videoconferencing over ATM: Business Quality
H.323 - Videoconferencing over IP/Ethernet: Best Effort Quality H.324 - Videoconferencing over POTS: Low Quality H.310 - Videoconferencing MPEG-2 over ATM: Broadcast Quality
Videoconferencing Standards
K. Salah 10
Videoconferencing Standards
K. Salah 11
Issues to addressIssues to address
When deploying such a network service, When deploying such a network service, network architects, managers, planners, network architects, managers, planners, designers, and engineers are faced with designers, and engineers are faced with common strategic, and sometimes challenging, common strategic, and sometimes challenging, questions. questions. What are the QoS requirements for videoconferencing? How will the new videoconferencing load impact the
QoS for currently running network services and applications?
Will my existing network support videoconferencing and satisfy the standardized QoS requirements?
If so, how many videoconferencing sessions can the network support before upgrading prematurely any part of the existing network hardware?
K. Salah 12
Commercial ToolsCommercial Tools EURESOM Jupitor II has a provision to test end-to-end Quality of Service EURESOM Jupitor II has a provision to test end-to-end Quality of Service
(QoS) for Network-QoS-aware applications over IP networks. (QoS) for Network-QoS-aware applications over IP networks. NetIQ’s Vivinet Assessor generates RTP streams to mimic VoIP traffic NetIQ’s Vivinet Assessor generates RTP streams to mimic VoIP traffic
between pairs of hosts and assesses the quality of these synthetic calls.between pairs of hosts and assesses the quality of these synthetic calls. BMC PATROL DashBoard analyzes the impact of multimedia services on BMC PATROL DashBoard analyzes the impact of multimedia services on
the existing network. This tool can quickly identify specific problems on the the existing network. This tool can quickly identify specific problems on the network that impact application performance. network that impact application performance.
Spirent’s IPTV system is a product that includes various features like video Spirent’s IPTV system is a product that includes various features like video infrastructure testing, IPTV video quality testing, firewall and video server infrastructure testing, IPTV video quality testing, firewall and video server load testing. load testing.
RADVISION offers tightly integrated infrastructure processing components RADVISION offers tightly integrated infrastructure processing components called viaIP, for desktop and meeting room conferencing. called viaIP, for desktop and meeting room conferencing.
Omegon, Lucent VitalSuite, ViDeNet. Omegon, Lucent VitalSuite, ViDeNet.
"H.323 Beacon" tool is a open-source tool for assessing performance of "H.323 Beacon" tool is a open-source tool for assessing performance of desktop videoconferencing sessions using H.323 traffic emulation.desktop videoconferencing sessions using H.323 traffic emulation.
K. Salah 13
Two common approaches in assessing the Two common approaches in assessing the deployment of videoconferencing into the deployment of videoconferencing into the existing network. existing network. One approach is based on first performing network
measurements and then predicting the network readiness for supporting videoconferencing. The prediction of the network readiness is based on assessing the health of network elements.
The second approach is based on injecting real videoconferencing traffic into existing network and measuring the resulting delay, jitter, and loss.
K. Salah 14
LimitationsLimitations
None of the commercial tools offers a None of the commercial tools offers a comprehensive approach for successful VoIP comprehensive approach for successful VoIP deployment. deployment.
In particular, none gives any prediction for the In particular, none gives any prediction for the total number of calls that can be supported by total number of calls that can be supported by the network taking into account important design the network taking into account important design and engineering factors. and engineering factors. These factors include VoIP flow and call distribution,
future growth capacity, performance thresholds, and impact background traffic.
This presentation attempts to address those This presentation attempts to address those important factors utilizing OPNET simulation. important factors utilizing OPNET simulation.
K. Salah 15
Case StudyCase Study
Floor 1
Internet
7
Switch 1 Switch 2
. . .
User PCs
Workgroupserver
Floor 3
. . .
User PCs
Floor 2
. . .
User PCs
Databaseserver E-Mail
server
Fileserver HTTP
server
Web &cacheproxy
Firewall
Router
Printerserver
Workgroupserver
Printerserver
Workgroupserver
Printerserver
K. Salah 16
Videoconferencing-Enabled IP NetworkVideoconferencing-Enabled IP Network
Floor 1
Internet
7
Switch 1 Switch 2
Floor 3Database
server E-Mailserver
Fileserver HTTP
server
Web & cacheproxy
Firewall
Router
Gatekeeper
. . .
Workgroup server
Printer server
Multimedia PCs
Floor 2
. . .
Workgroup server
Multimedia PCs
Printer server
. . .
Workgroup server
Multimedia PCs
Printer server
K. Salah 17
At minimumAt minimum
H.323H.323 gatekeepergatekeeper handles signaling for establishing, terminating, and
authorizing connections of video sessions, as well as imposing maximum bandwidth for each session.
H.323 workstations or multimedia PCsH.323 workstations or multimedia PCs equipped with H.323 voice and video software equipped with a camera and a microphone.
Switched LANSwitched LAN No need for MCU No need for MCU
Multipoint Control Unit supports conferencing among three or more endpoints
We assume p2p Desktop VC only
K. Salah 18
Traffic Flow and Call Distribution Traffic Flow and Call Distribution
Traffic flow has to do with the Traffic flow has to do with the path that session travels path that session travels through. through.
Session distribution has to do Session distribution has to do with the percentage of with the percentage of sessions to be established sessions to be established within and outside of a floor, within and outside of a floor, building, or department. building, or department.
The intra-floor traffic will The intra-floor traffic will constitute 20% of over all constitute 20% of over all traffic, and the other 80% will traffic, and the other 80% will constitute inter-floor traffic. constitute inter-floor traffic. Such a distribution can be
described in a simple probability tree as shown.
Total calls
F1 F2 F3
1/31/31/3
F1 F2 F1 F3 F2 F3
Intra-floor Inter-floor
4/51/5
1/31/31/3
K. Salah 19
Additional ConsiderationsAdditional Considerations
We assume voice and video calls are symmetric. We assume voice and video calls are symmetric. we assume a point-to-point desktop videoconferencing.we assume a point-to-point desktop videoconferencing.
Streaming stored video and broadcast video is not considered in this presentation.
We also ignore the signaling traffic generated by the We also ignore the signaling traffic generated by the gatekeepergatekeeper. . We consider the worst-case scenario for videoconferencing traffic.We consider the worst-case scenario for videoconferencing traffic. The signaling traffic involving the gatekeeper is only generated prior to
the establishment of the session and when the session is finished. This traffic is relatively limited and small compared to the actual voice call traffic.
In general, the gatekeeper generates no signaling traffic throughout the duration of the videoconferencing session for an already established on-going session.
In order to allow for future growth, we will consider a 25% growth In order to allow for future growth, we will consider a 25% growth factor for all network elements including router, switches, and links. factor for all network elements including router, switches, and links.
K. Salah 20
Delay and Bandwidth RequirementsDelay and Bandwidth Requirements
The actual number of videoconferencing The actual number of videoconferencing sessions that a given network can sustain sessions that a given network can sustain and support is bounded by and support is bounded by End-to-end delay Bandwidth
Either the available bandwidth or delay Either the available bandwidth or delay can be the key dominant factor in can be the key dominant factor in determining the number of sessions that determining the number of sessions that can be supported. can be supported.
K. Salah 21
BandwidthBandwidth
A videoconference session consists of two independent A videoconference session consists of two independent bidirectional streams: voice and videobidirectional streams: voice and video
For voice, the required bandwidth for a voice call on any For voice, the required bandwidth for a voice call on any one direction, is 50 pps or 90.4 kbps with packet one direction, is 50 pps or 90.4 kbps with packet overhead. For both directions, the required bandwidth for overhead. For both directions, the required bandwidth for a single call is 100 pps or 180.8 kbps assuming a a single call is 100 pps or 180.8 kbps assuming a symmetric flow. symmetric flow. Packet size is fixed at 160 bytes
For video, the packet sizes for video traffic are variable. For video, the packet sizes for video traffic are variable. variability in the packet sizes depends on the actual temporal
and spatial nature of the video content being encoded. Typically, one video frame is packetized in one Ethernet frame
with sizes ranging from 65-1518 bytes.
K. Salah 22
Video packet size distributionVideo packet size distribution
Characteristics collected from well-known Characteristics collected from well-known Internet testbed Internet testbed
Packet sizes orrespond to an aggregated Packet sizes orrespond to an aggregated representation of video traffic from H.261, representation of video traffic from H.261, H.262 and H.263 video codec streams H.262 and H.263 video codec streams arising from desktop videoconferencing arising from desktop videoconferencing end-points end-points
K. Salah 23
Histogram and CDFHistogram and CDF
K. Salah 24
Two ScenariosTwo Scenarios FixedFixed
Video packet sizes of 1344 bytes, and sent at a rate of 30 fps The range of packet sizes above 512 bytes constitute close to 65% of all
packet sizes. This gives approximately a rate of 320 kbps for pure video traffic.
A bandwidth of 320kbps is a multiple of the basic 64kbps communication channel and is an acceptable bandwidth for business desktop videoconferencing with default recommendations of H.261 video codec, CIF video resolution, and H.323 frame rate.
When considering the additional 66 bytes of layer headers, similar to byte overhead for VoIP, the required bandwidth for a video call would be 338.4 kbps.
For both directions, the required bandwidth for a single video call is 60 pps or 676.8 kbps assuming a symmetric flow.
Hence, for a bidirectional videoconferencing session the required bandwidth is 160 pps or 857.6 kbps.
EmpiricalEmpirical Actual packet sizes are imported, and sent at a rate of 30 fps
K. Salah 25
End-to-end DelayEnd-to-end Delay
According to recommendations by ITU, when delays are less than According to recommendations by ITU, when delays are less than 150 ms, most interactive applications, both speech and non-speech, 150 ms, most interactive applications, both speech and non-speech, will experience essentially transparent interactivity. will experience essentially transparent interactivity.
For voice, the end-to-end delay is sometimes referred to by M2E or For voice, the end-to-end delay is sometimes referred to by M2E or Mouth-to-Ear delay should be less than 150 ms. Mouth-to-Ear delay should be less than 150 ms.
In videoconferencing, there is no separate delay for voice and video In videoconferencing, there is no separate delay for voice and video streams as both voice and video are synchronized in what is streams as both voice and video are synchronized in what is commonly known as “lip-sync”. commonly known as “lip-sync”.
According to real experimental work, the delay difference (termed According to real experimental work, the delay difference (termed also skew) between voice and video should be less than 80 ms to also skew) between voice and video should be less than 80 ms to allow for natural human interaction and impression. allow for natural human interaction and impression.
For our upper bound end-to-end one-way delay of a video or voice For our upper bound end-to-end one-way delay of a video or voice packet, we will use 100 ms. This can be broken into 80 ms for the packet, we will use 100 ms. This can be broken into 80 ms for the network delay and 20 ms delays for both sender and receiver network delay and 20 ms delays for both sender and receiver workstations. workstations.
K. Salah 26
Simulation StudySimulation Study
K. Salah 27
Generating Videoconferencing Traffic Generating Videoconferencing Traffic
K. Salah 28
Voice and video profile settingsVoice and video profile settings
K. Salah 29
Settings of multimedia workstationsSettings of multimedia workstations
K. Salah 30
Simulation ResultsSimulation Results
Examine jumps in pps Examine jumps in pps for voice and videofor voice and video Voice – 300 pps Video – 180 pps
Examine jumps in Examine jumps in bytes/secbytes/sec Voice – 67.8 K bytes/s Video – 253.8 K bytes/s
K. Salah 31
Global videoconferencing traffic in pps Global videoconferencing traffic in pps
Since the last successful Since the last successful addition point was the addition point was the same for voice and video same for voice and video (at 2:56), this yields to (at 2:56), this yields to videoconferencing calls ofvideoconferencing calls of
Also examine the Y axisAlso examine the Y axis Video -- divide by 60 Voice – divide by 100
1623)2/)7056602((3
K. Salah 32
Global videoconferencing end-to-end delayGlobal videoconferencing end-to-end delay
K. Salah 33
RouterRouter
K. Salah 34
K. Salah 35
Utilization of Router Utilization of Router Switch links Switch links
K. Salah 36
Empirical Video PacketsEmpirical Video Packets
OPNET can be configured to OPNET can be configured to use empirical video packets by use empirical video packets by simply changing the values of simply changing the values of incoming and outgoing stream incoming and outgoing stream frame size to OPNET special frame size to OPNET special “scripted” distribution in which “scripted” distribution in which a filename containing the a filename containing the packet sizes is specified. packet sizes is specified.
In the figure “pkts” is the filename
K. Salah 37
Simulation Results for Empirical Video Simulation Results for Empirical Video PacketsPackets
K. Salah 38
A stable simulation run of 144 sessionsA stable simulation run of 144 sessions
K. Salah 39
Concluding RemarksConcluding Remarks
With similar analysis, the total With similar analysis, the total videoconferencing sessions to be videoconferencing sessions to be supported is 162. supported is 162.
A successful simulation run (with no A successful simulation run (with no packet loss and a delay of 22.5 ms) of 144 packet loss and a delay of 22.5 ms) of 144 videoconferencing sessions were obtainedvideoconferencing sessions were obtained for fixed video packet sizes for empirical video packet sizes
Did not make a difference in a stable runDid not make a difference in a stable run