Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Peer-To-peer Architecture Could Save Streaming Video

webrtcworld.com

Post reply on HN
Print view View original

11–20 of 37 posts

Re: Peer-To-peer Architecture Could Save Streaming Video

#11

Post by basiliothecat »

jasode wrote: post #6

>Instead of one endpoint pulling down a single unique stream, the P2P approach allows simultaneous users to exchange video segments among themselves rather than each connecting to a server to do so. Issues: -- residential/mobile internet service has asymmetric upload/download speeds. -- ISPs are suspicious of heavy uploads originating from residential homes and could be flagged as "servers" in violation of ISP's TOS…

The presence of those consumer/server distinction in ISP's TOS should get waded out at some point. And distributed content is much move viable solution than a centralized ones - probably a troublesome one regarding streaming live data, but one which is potent to utilize network for the task at hand more harmoincally. And at the same time it'll bring all the benefits of distributed systems (as well as downsides) - e.g. you could stream a video of yours on your own website and not rely on likes of youtube.

Re: Peer-To-peer Architecture Could Save Streaming Video

#12

Post by joncrocks »

jasode wrote: post #6

>Instead of one endpoint pulling down a single unique stream, the P2P approach allows simultaneous users to exchange video segments among themselves rather than each connecting to a server to do so. Issues: -- residential/mobile internet service has asymmetric upload/download speeds. -- ISPs are suspicious of heavy uploads originating from residential homes and could be flagged as "servers" in violation of ISP's TOS…

In terms of live events this was meant to be solved by multicasting.

http://en.wikipedia.org/wiki/IP_multicast or http://en.wikipedia.org/wiki/Broadcast_address

The point being that if you have a stream of data that is the same for a large number of recipients, you can send the data down the tubes once, and it's then shared by recipients.

Although doesn't solve the general case (people wanting to watch the same thing, but out of order).

Re: Peer-To-peer Architecture Could Save Streaming Video

#13

Post by rubiquity »

dexen wrote: post #4

A whole article on p2p video streaming, and yet Popcorn Time [1] is curiously absent. The flagship program that proves p2p movie watching is feasible, even comfortable, when delivered through BitTorrent doesn't get a mention. In a way the program already saved streaming video already. We can only speculate whether the omission is conscious attempt at avoiding linking to risque programs, or does that indicate the arti…

That is probably due to this article being about WebRTC and not Torrents.

Re: Peer-To-peer Architecture Could Save Streaming Video

#15

Post by KaiserPro »

Sorry, but no.

For a p2p system to work you need to have a nearest neighbour finder(please wait 5 minutes whilst we probe your network) you then need to have a realtime "rationilser" that will go through and re-route the live stream according to how many peers in one segment are connected to another. Then, you need a number of backup peers for when your current master peer disappears.

Then you have to deal with partition, and also make sure that NAT doesn't get in the way. Also you'll need to block mobile phones as well.

the best part is, you'll be forced to use HVEC as the usable upload is around 2 megabits. That means a stream encoded at 750k (to allow uploading to two other users) or half that if you want 4 other peers.

So no, P2P is not the answer. Iterate again.

Re: Peer-To-peer Architecture Could Save Streaming Video

#16

Post by diydsp »

Note that this article claiming online video streaming needs to be saved is on a WebRTC marketing site "Powered by Technology Marketing Corp" advocating a competing technology.

I don't think online video streaming needs to be saved.

> There’s a clear event horizon where delivery overhead outstrips even CDN capacity.

If it's so clear, please enlighten me: All I see is network architecture making ever bolder jumps to increase capacity... Look at Amazon's multiple locations, look at fiber-to-the-home, look at the server farms around the country. Look at the bold instantiation of cell towers becoming ever-denser.

I think if the public demands 150 million individual 10Mbps streams, it will be built.

Not that WebRTC isn't cool for what it is, but a soccer game isn't a volcano (10 people punting a ball/data around like WebRTC isn't the same as a massive one-way structure of energy/data like a Volcano).

Re: Peer-To-peer Architecture Could Save Streaming Video

#18

Post by sylvinus »

KaiserPro wrote: post #15

Sorry, but no. For a p2p system to work you need to have a nearest neighbour finder(please wait 5 minutes whilst we probe your network) you then need to have a realtime "rationilser" that will go through and re-route the live stream according to how many peers in one segment are connected to another. Then, you need a number of backup peers for when your current master peer disappears. Then you have to deal with parti…

Did you read the article? Your points are irrelevant because it's not 100% P2P, they say they reach ~58% of P2P use, while the rest comes from regular CDNs.

Re: Peer-To-peer Architecture Could Save Streaming Video

#20

Post by avn2109 »

jasode wrote: post #6

>Instead of one endpoint pulling down a single unique stream, the P2P approach allows simultaneous users to exchange video segments among themselves rather than each connecting to a server to do so. Issues: -- residential/mobile internet service has asymmetric upload/download speeds. -- ISPs are suspicious of heavy uploads originating from residential homes and could be flagged as "servers" in violation of ISP's TOS…

Doesn't Spotify do p2p for fairly-high-resolution audio already?
Post reply on HN