Re: [httpstreaming] Push Vs Pull (was Re: Current Status and Our Goal)

All,

Now that it is clearer what technology some of the participants here would like to consider (namely the so-called "push model"), I think it would be reasonable to ask what specific standards they think are missing to implement and deploy this technology ? What work should the IETF do ? Is it proposed to restrict the scope to this "push model" ? (which might be reasonable, since I do not think additional standards work is required in IETF to support so-called "pull model").

Whatever work is proposed it is worth noting that the industry trend is very much towards the "pull model". Far more media is being delivered in an adaptive way with that model than with the "push model". This point might be considered when priortising this work.

Neverthless, for the push model there are already multiple standards for media delivery over a single HTTP connection and multiple deployed products which do this, including adaptation of the video rate, by transcoding, switching or otherwise. You could look at Adobe Flash for example. For real-time adaptation from a single source you could look at Ortiva or Vantrix. I'm sure there are others.

...Mark

On Oct 15, 2010, at 5:02 PM, Qin Wu wrote:

> Thank you for separate this issue out.
> ----- Original Message ----- 
> From: "Ben Niven-Jenkins" <ben@niven-jenkins.co.uk>
> To: "Qin Wu" <sunseawq@huawei.com>
> Cc: <httpstreaming@ietf.org>
> Sent: Friday, October 15, 2010 4:57 PM
> Subject: Push Vs Pull (was Re: [httpstreaming] Current Status and Our Goal)
> 
> 
> 
> On 15 Oct 2010, at 08:14, Qin Wu wrote:
> 
>> From: "Ali C. Begen (abegen)" <abegen@cisco.com>
>> 
>>> But, in http streaming scenario, the ratio of download/upload (from the client's perspective) is much larger than 1. BTW, if
>>> you keep your chunk duration relatively longer, the amount of requests that you will end up sending will be almost nil
>>> compared to what you will receive.
>>> 
>>> [Qin]: No, I just compare pull and push with the same chunk duration.
>> 
>> You can choose whatever chunk size you wanna use. Does not matter. The fact remains the same. Unless someone uses chunks of a few hundreds of ms, it won't matter.
>> 
>> [Qin]: Suppose 10 chunks is available at the server side, in the pull model, the client need to send at least 10 requests and receive 10 responses.
>>         in the push model, the client may only need to send one request and then receive 10 responses.
>>         comparing with pull, push model save 9 requests.
>>         also if you look at websocket, push has more lightweight header than pull.
> 
> However, unless the request/response time or the volume of requests per server is a significant bottleneck in the system then saving some requests doesn't provide any advantages.
> 
> [Qin]: Good point. Such bottleneck could happen in case of concurrence of thousands of consumers accessing the same content. 
> 
> Ben
> _______________________________________________
> httpstreaming mailing list
> httpstreaming@ietf.org
> https://www.ietf.org/mailman/listinfo/httpstreaming
> 