cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1651
Views
5
Helpful
4
Replies

Dedicated Bandwidth to Endpoints

Saurabh Gupta
Level 3
Level 3

Hi All,

I have one question :- How can I refrain from providing dedicated bandwidth to my endpoints in entire network infrastructure , For eg , HD call needs minimum 2 Mbps for a good call. So, if I provided Dedicated 2 Mbps bandwidth to endpoint , it is of no use as I am reserving bandwidth to one endpoint permanently.

Is there anyway , where I Can allocate Bandwidth to an endpoint in an adhoc way where and when needed so that It doesnt reserves the bandwidth at all time.?

Look forward for your valuable suggestions.

Regards,

Saurabh Gupta

2 Accepted Solutions

Accepted Solutions

Saurabh, your endpoints don't use bandwidth when they're not in a call.  There may be a very small amount of RAS traffic, but this is very small and will go unnoticed by other devices on your network.

Typically most customers either don't worry about reserving bandwidth for video, or they use QoS mechanisms on their network to make sure that the video (which is real-time, and sensitive to delays and loss) isn't impacted by other data needs.

This is different from dedicated bandwidth, where the bw is set aside all the time for just video.  QoS kicks in to make sure video has the bw it needs when your network starts to get congested.

I've personally never seen anyone try to dedicate a portion of their network bw exclusively for video.  If they're going that far, then they typically have a completely separate network for video.

View solution in original post

antigles
Cisco Employee
Cisco Employee

Hi Saurabh,

You will typically enable QOS end to end across your network. Some customers deploy a dedicated network for TelePresence and many others, like you, use their existing network and share the same infrastructure for all data, voice and video applications.

The solution will always rely on applying a DiffServ model for QoS where every network devices recognize traffic classes and provide different levels of QoS todifferent traffic classes.

On your WAN you can, for instance, configure LLQ assigning TP traffic a prioritary treatment over your data. Unused TP bandwidth will be used by data when not in a TP call. Assigning TP traffic to a class of CBWFQ is also possible.

The literature about this is very extensive, but it is very well explained and tailored around TelePresence in the following document:

Cisco TelePresence Network Systems 2.0 Design Guide http://www.cisco.com/en/US/docs/solutions/Enterprise/Video/TP-Book.html

You will find there explanations on how technology works, on how to plan and design QOS on your network and it includes specific examples on how to configure all incumbent LAN and WAN network devices.

Best regards.

View solution in original post

4 Replies 4

mubakhta
Level 4
Level 4

Hi Saurabh,

How exactly are you trying to dedicate the bandwidth to the endpoint? Are you using the Max Transmit and Max Receive in order to reserve the bandwidth. You can definitely define the Maximum bandwidth using the Max Transmit and Max Receive settings.

Once you have set the Max Transmit Bandwidth, it is then distributed among the Main Channel and Presentation channel. You can refer to this documentation piece :

xConfiguration Conference [1..1] VideoBandwidth Mode

Set the conference video bandwidth mode.

Requires user role: ADMIN

Value space:

Dynamic: The available transmit bandwidth for the video channels are distributed among the

currently active channels. If there is no presentation, the main video channels will use the

bandwidth of the presentation channel.

Static: The available transmit bandwidth is assigned to each video channel, even if it is not active

Hence, if you then decide to distribute, you can distribute it accordingly to be dynamic or you can have it static which would allow you to reserve the bandwidth for the Main as well as Presentation Channels.

Hope this answers your question.

Regards,

Mubashshir Akhtar

Thanks Mubashshir Akhtar

Hi Mubashshir,

Thanks for replying. Glad to See you doing well.!!

However, the question I asked above is slightly different, I am asking " Do we supposed to give dedicated bandwidth to endpoints for good HD Call " or our endpoints are intelligent enough to take desired bandwidth from network once it wants to make call and then give the bandwidth back to Network once it is done ?

What is the common practice followed by other Cisco Customers? If yes, please tell me the way to do it.

Thanks,

Saurabh

Saurabh, your endpoints don't use bandwidth when they're not in a call.  There may be a very small amount of RAS traffic, but this is very small and will go unnoticed by other devices on your network.

Typically most customers either don't worry about reserving bandwidth for video, or they use QoS mechanisms on their network to make sure that the video (which is real-time, and sensitive to delays and loss) isn't impacted by other data needs.

This is different from dedicated bandwidth, where the bw is set aside all the time for just video.  QoS kicks in to make sure video has the bw it needs when your network starts to get congested.

I've personally never seen anyone try to dedicate a portion of their network bw exclusively for video.  If they're going that far, then they typically have a completely separate network for video.

antigles
Cisco Employee
Cisco Employee

Hi Saurabh,

You will typically enable QOS end to end across your network. Some customers deploy a dedicated network for TelePresence and many others, like you, use their existing network and share the same infrastructure for all data, voice and video applications.

The solution will always rely on applying a DiffServ model for QoS where every network devices recognize traffic classes and provide different levels of QoS todifferent traffic classes.

On your WAN you can, for instance, configure LLQ assigning TP traffic a prioritary treatment over your data. Unused TP bandwidth will be used by data when not in a TP call. Assigning TP traffic to a class of CBWFQ is also possible.

The literature about this is very extensive, but it is very well explained and tailored around TelePresence in the following document:

Cisco TelePresence Network Systems 2.0 Design Guide http://www.cisco.com/en/US/docs/solutions/Enterprise/Video/TP-Book.html

You will find there explanations on how technology works, on how to plan and design QOS on your network and it includes specific examples on how to configure all incumbent LAN and WAN network devices.

Best regards.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: