January 25, 2007 | Posted By : Isabel Wang at 6:29 PM PERMALINK | Blog Comments (10)

Apple Could Have Saved $15 Million if iTunes were P2P?

That's what this very funny blog post from RedS Swoosh says. I came across it through Read/Write Web, where Richard McManus points out that Mark Cuban of the Dallas Mavericks is a Red Swoosh investor.

Red Swoosh is a free P2P CDN. Content owners "swoosh" their content by adding "http://edn.redswoosh.net" in front of URLs. Visitors must install a desktop client in order to access swooshed data. Files are downloaded or streamed from up to 30 peers who previously requested the same content.

Johnny Cakes, Red Swoosh's probably non-existent "Downloader in Chief", read on the BBC that Apple has sold 1.5 billion songs and tens of millions of TV shows and movies through iTunes. He calculates that...

* 1.5 billion songs x 5 MB = 7,500 TB
* 25 million TV shows x 700 MB = 17,500 TB
* 25 million movies x 2 GB = 50,000 TB
* 75,000 TB x $200 per TB in Akamai bandwidth = $15 million, which Apple could use to:

* Buy the world a Coke
* Give away 100,000 iPhones
* Buy 3,750 Segways and a lifetime supply of hot dogs
* Buy Sealand

Red Swoosh is not the only P2P CDN. The New York University-operated CoralCDN is free as well (and doesn't require a desktop client); Fark, Slashdot and Digg are all occasional users. The Coral team has been collecting data on the distance between web users and their DNS resolvers to determine how efficient it is for CDNs to make IP-address-based matches between end users and cache servers.

There's also Metalink, an open standard that bundles P2P files AND HTTP/FTP files on multiple mirrors into one single format. If a server becomes inaccessible during a download, Metalinker will automatically switch to a different location. It's even able to download segments from different sources at the same time. OpenOffice and openSUSE (among many other applications) are available via Metalink.

If you have hosting customers who need more bandwidth than they can afford, work with them to evaluate these options instead of canceling their accounts. If enough high-bandwidth users start offering P2P content, it might buy you some time before you have to fire up the next batch of 10-GigEs.

Print | Send | del.icio.us | Linking Blogs

Comments

[Add Comment]

The big problem with P2P CDN's is that the require folks to have an active uploader. The problem with RedSwoosh, in particular, is that this active uploader runs in the background, is difficult to uninstall, and looks like spyware. And, of course, every different P2P CDN needs a different active uploader. Thats the key reason why Akamai and Limelight are not worried about these P2P startups.

# Posted By Daniel Golding | 1/26/07 8:15 AM

Dan, CoralCDN does NOT require folks to have an active uploader. Also, I didn't mention VeriSign's Kontiki, since it's P2P but not free. I think Kontiki's recent deal with Adobe does carry some weight; it seems like something Akamai and Limelight should keep an eye on.

# Posted By Isabel Wang | 1/26/07 8:57 AM

Thats because CoralCDN isn't really P2P - at least, not yet. All of the server/cache boxes on Coral are under NYU's control. They are utilizing the PlanetLabs research consortium. Think of Coral more like an open-source version of Akamai - highly distributed, but not really P2P. Its free now because its paid for by grants. It could be free in the future, but folks would need to contribute bandwidth and set up servers.

Sadly in life, nothing is free. A CDN can pay for transit/peering bandwidth or can use their user's upstream bandwidth (requiring a client or browser plugin). Kontiki, of course, does have a client.

From what I've heard, the primary interest in P2P from Akamai and Limelight is that for large file downloads, they do sometimes utilize download manager clients (to enable users to restart transfers, etc). It would be possible to put P2P client functionality into those download managers so that, while a user's download is going on, some of their upstream bandwidth could be utilized to accelerate the overall content distribution process. I am envisioning a "turbo" button in the client to enable upstream (and faster downstream) data transfer. But that is an ephemeral download manager, rather than a persistent client.

# Posted By Daniel Golding | 1/26/07 9:29 AM

Ok, I see what you're saying. Coral's DNS layer is maintained by NYU, so while users are able to perform P2P data transfers, their ability to find each other is dependent on Coral's infrastructure.

So Coral seems to represent a 3rd approach to the two alternatives you mentioned? I'm not sure Coral itself caches any content. And speaking of The PlanetLabs research consortium, I was looking on their membership list, and Google is on there.

As for your "turbo" functionality, it sounds a lot like what Metalink does - except Metalink is a persistent client. I really like their approach of enabling simultaneous downloads of different file segments from multiple P2P sources.

# Posted By Isabel Wang | 1/26/07 10:13 AM

Isabel,
Despite other errors in the redswoosh post, why not take official figures from Apple? There are also errors in the math which I see, and I'm horrible at math.

http://www.apple.com/pr/library/2007/01/09itunes.h...
http://docs.info.apple.com/article.html?artnum=304...

2 billion songs
50 million TV shows
1.3 million movies

"How large is the average movie?

A 2-hour movie is about 1.5 GB. "

I do agree that P2P can lead to great bandwidth savings. It would be interesting to see if a major app like iTunes could pull it off. Fine for my computer at home, but not on LANs (universities or offices) or places w/ bw caps.

Thanks for mentioning Metalink. You can also give priority to certain resources (P2P or mirrors) so you could say "Use BitTorrent first, then magnet links, then HTTP as a last resort if P2P is blocked."

# Posted By Ant Bryan | 1/26/07 4:21 PM

Ok, let's redo Red Swoosh's math:

2 billion songs x 5 MB = 8,000 TB
50 million TV shows x 700 MB = 35,000 TB
1.3 million movies x 1.5 GB = 1,950 TB

That's about 50,000 TB instead of 75,000, so Apple would only save $10 million.

BTW, my favorite part of your site has always been how "438,784 llamas agree that Metalink improves their quality of life" :)

# Posted By Isabel Wang | 1/26/07 6:12 PM

While I can understand the point of saving money, the eventual cost savings are going to add up likely in lost customers. I feel that something that iTunes users greatly enjoy is the fact that their upload bandwidth isn't shared when they're logged in, allowing them to still browse the internet at just a slightly delayed speed with most broadband connections. The second area is, as we all know, P2P isn't the fastest when downloading, and that's another area with downloading from a central server connected to some huge pipes has a huge benefit. I max out my DSL Connection every time when downloading from iTunes - I rarely max out my downstream when downloading from P2P.

# Posted By Terry Myers | 2/5/07 8:22 PM

Terry - Your iTunes files most likely aren't coming through huge pipes connected to a central server. iTunes currently works with Akamai, which operates 20,000+ cache servers in 70+ countries.

CDNs reduce content owners' bandwidth consumption by minimizing the distance between end users and their requested data. P2P CDNs have the potential of bringing content even closer to you. Technologies like Metalink, in particular, support simultaneously downloads of different file segments from multiple sources.

While you may not be all that impressed with your P2P experience to date, think about how Internet access has evolved over time. I still remember having 14.4K dialup; P2P would have been completely impossible back then. Now you have DSL; P2P is more feasible but still not that great. But imagine a similarly significant improvement in your home connectivity 3, 5 or 10 years from now. Can you see a point where P2P starts to make a lot of sense?

# Posted By Isabel Wang | 2/5/07 11:24 PM

I too remember the days of 14.4 dialup, although at that time at the age of 10 I had absolutely no clue what it all meant. I'm not actually sure how I had the patience to sit through the slow loading times back then. You do however raise an excellent point about the future growth of P2P. When you look back to when P2P first became mainstream with Napster, and then how it "upgraded" to Kazaa, and now it's with the torrents, it's come a long ways. It's not only faster, but it's much more reliable.

RedSwoosh looks really interesting, especially for the average web master who's running a fan site, or even an advertising free site that burns through bandwidth. It could definitely lower the costs of some bandwidth intensive files, whether it's a clip of a Hank Aaron home run, Dale Earnhardt winning the Daytona 500 or something random clip. The fact that it requires a separate client though really puts a kink in it since you're asking the end user to really cater to you rather than you catering to them as usual.

Back to iTunes though. It would have been excellent if Apple were able to save that kind of money, and instead use for some philanthropic or lowering of an Apple PC, or whatever, but with iTunes and the iPod being a "high end" product, or at least that's what Apple likes to market it as, I feel that until they can guarantee the high speed download rates, P2P just isn't feasible. I could just see my friends saying "iTunes sucks, it's too slow to download songs. Just get the Zune!" because it took them 3 minutes to download their favorite new song, rather than the 30 seconds it'd take running the CDN style system that they're using now.

It is however a very interesting thought. Smaller companies who aren't able to afford the costs of Akamai's CDN Product should definitely take a look at these new P2P systems.

# Posted By Terry Myers | 2/6/07 1:01 AM

I'm not sure what the hold-up is... maybe they have re-thought their stance on how this is going to actually make the company any money. Or perhaps their lawyers pointed out the liability of providing agents a platform to stick their feet in their mouth. Whatever it is, it's hardly something I'd claim as being "Well done".
http://www.jebshouse.com/wordletter.php?l=G

# Posted By Jeb Simons | 5/18/08 3:45 PM

[Add Comment]