Any mention of Deep Packet Inspection (DPI) these days is bound to generate heated discussion. Everybody in our industry has a strong position, either viewing peer-to-peer (P2P) filtering as mandatory and the only way to manage consumer traffic in Internet service provider (ISP) networks and/or to enforce copyright law, or as evil because a service provider shall treat all customers equally and the media industry should accept the end of its business model.
In the end, we found that mentioning money quieted opponents and made the business case for ISPs obvious. None of the P2P users we met (in a non-representative internal survey) was ready to pay for premium, 24x7, wire-speed, broadband Internet access. And few service providers seem to be ready to spend money on a huge, unbeatable consumer network. But protocol filtering to manage the vast amounts of data from P2P power users (more than 50 percent of all Internet traffic from less than 10 percent of the users, according to one study) is the only way to achieve both goals today. Quod erat demonstrandum. Three protocol filtering vendors participated in the test and showed off great results.
We were not so sure about the content filtering part of the story. The media industry keeps fighting, specifically in European countries. The recent Pirate Bay trial and the discussion going in France about the HADOPI law ("HADOPI" being the acronym for the government agency of the Haute Autorité pour la Diffusion des Œuvres et la Protection des droits sur Internet created by this bill), which regulates and controls Internet usage in order to enforce compliance with copyright laws, are only the most prominent examples. It is also interesting to monitor what has happened in countries where courts already ruled in favor of the media industry a while ago, like Ireland and Belgium.
Unlike the March 2008 test, which was paid for by SNEP, the French recording industry association, each vendor in this latest test paid to participate. Vendors had only a generic veto right to opt out, not specific editing rights related to individual test cases. Vendors got the right to review their results and our interpretations related to their results; we did not share the full report with them, nor the results or names of the other participating vendors. And in contrast to our first test, we also increased the testbed scale to 25 Gbit/s at the P2P application layer and added the latest P2P protocols.
Unfortunately, not a single content filter vendor was ready to publish its results in our test. Why? According to our sources, some pretty amazing functionality has been implemented, but it just doesn't scale yet. Audible Magic rejected participation in our test, unfortunately. Other vendors, like SafeMedia Corp. , had publicly expressed interest to join our previous test, but when we invited them to the current campaign, nobody responded. Yet another content filter vendor expressed interest, but then claimed that all its resources were completely consumed by a trial in Australia (although we patiently waited for more than three months).
Therefore, we cannot report that media content filtering solutions (as opposed to protocol-based filtering) would be ready for today’s scale of broadband Internet networks based on our testing.
— Carsten Rossenhövel is Managing Director of the European Advanced Networking Test Center AG (EANTC), an independent test lab in Berlin. EANTC offers vendor-neutral network test facilities for manufacturers, service providers, and enterprises. Carsten heads EANTC's manufacturer testing and certification group and is responsible for the design of test methods and applications.
I don't think the ISPs are concerned about illegal content going through their networks (as in moral issues). They are concerned that traffic is going through their networks and they are not earning anything on it.
P2P is today's headache because there's soooo much traffic but ISPs just want people to pay for their broadband connection and use it as little as possible. I am sure they have issues with YouTube streaming videos for free and if NetFlix continues to grow, I'm sure they'll have an issue with them as well.
There's merit to what you're saying. However, it's not entirely a matter of "whether" pirates will change and why. There are many P2P technologies, today, and new ones crop up frequently, wavering at varying levels of adoption. For the most part, there's a high-school popularity content forever pushing-and-pulling different "solutions." There is a lot of cross-pollination, too. While most file-sharers have a preferred technology, many will expose their folder to several technologies at once -- resulting in a cross-pollinating P2P technology mesh. Such people might scarcely notice that one of their channels is snuffed out, and it's a double-click install for their "friends" to continue getting stuff.
These things are as likely to change for arbitrary reasons as they are due to real technology pressures.
The only realistic way to curb it is via bandwidth limits. However, that threatens services like Netflicks and products like Roku (http://www.roku.com/) which deliver media via the web in a legitimate fashion.
You AUP may say that you are not to put a server on-line using your home service broad-band connection
if you put p2p on your computer and then leave the computer up all the time so others can download from your share you have put a server on line, possibly violating your aup
Very good point. That is exactly what I kept thinking for a while: Isn't all this filtering a fake idea if a bunch of Linux geeks can make it useless in weeks?
The answer is in understanding the dynamics of collective decision making. The masses don't change their systems until a substantial fraction (well-connected to the rest of the group) is utterly upset, or if people are actually forced to change. Once that state has been reached, a waterfall change can take place - but hardly before. I apologize for not quoting a proper scientific reference here; failed to lookup the articles electronically that I have read a long time ago.
I believe the way the smarter service providers use P2P filtering technology today reflects this theory:
1) Avoid upsetting the majority of users too much: Throttle P2P traffic where necessary but not too obvious, dumb or draconian. (Side note: This is why there is competition between filtering vendors to filter as smart as possible, sometimes even involving advanced connection admission control.)
2) Never force anybody to change their behavior radically: Do not, ever, filter any P2P protocols completely.
The first could start group dynamics going, the latter would force all users who want to continue using P2P to switch to a different program or method.
There are technical reasons why people may prefer not to encrypt as well - it can be slower depending on CPU speed vs. bandwidth available - but I believe the technical reasons are secondary.
Take away music, tv shows, movies, audio books, and software from my internet connection, and I suddenly have very little use for high-bandwidth internet access. "Download movies, tv shows, and music faster than ever before" has been the primary slogan used by internet providers to push peolpe to broadband for years. For a reason. Once we set aside issues regarding viability of a crippled product and falsified advertising, We are still ignoring some simple technology facts.
P2P traffic can travel on any port and it can be encrypted. It would take rudimentary programming skills to alter an open-source P2P client to send & receive traffic on a web port and encrypt it (for instance with ssl). With "average" technology on the comptuer end, bandwidth (not CPU) is still the bottleneck in an encrypt-transmit-decrypt P2P process when compared to a transmit-only technology.
P2P traffic is not being disguised as anything else at all right now, but that's simply because it isn't necessary. It would be easy as pie to disguise it as VPN traffic or encrypted web traffic or something else, and filtering any kind of encrypted P2P data would be almost impossible.
Assuming that ISPs spent any amount of money on packet-filtering technology, the hardware would be useless within the span of weeks or months.
The one of many tests you are talking about was not "bypass filtering" but rather protocol blocking. Procera allowed the most traffic while ipoque the least and cisco fell in between. The other tests such as total brandwidth Cisco totally failed. The Procera signature file was a year old so maybe that was the cause. What happen to Allot and Sandvine and a have dozen others?
I don't know that Cisco "lost," but it sure didn't shine, between the capacity/interface issue and the fact that ipoque allowed less than 0.01 percent of P2P traffic to bypass its filtering, compared with 2.4 percent for Cisco and 2.9 percent for Procera, as Carsten reported.
Horrible? No. Tragic? Hardly.
Cisco's never been a company to be first to market or to push the leading edge. It trails, then acquires. Like most market leaders (and I gang EMC, Microsoft, and Intel in this same category), Cisco is almost always the most expensive product on the shelf. But they know how to build relationships with customers and have been spectacularly successful at binding their customers to them.
Is a truck better than a station wagon because it is bigger? I guess it depends on what you want to achieve. To compare trucks and station wagons one should rather ask a different question: Which car fits its purpose best - in terms of being versatile, comfortable, efficient, and durable?
Our test aimed to validate those criteria in an unbiased way. Beyond the sheer throughput, it was most important for us to assess how many of the P2P protocols (traffic types) the devices would actually be able to identify and to filter. This technology is most difficult to develop. Packet filtering and forwarding hardware, on the other hand, accelerates like crazy. We witnessed a 25-fold increase in throughput within 15 months!
The level of precision and consistent performance that all three participating vendors showed was great and cannot be taken for granted across the whole market from my experience.
I was surprised to learn that Cisco's DPI device was half the capacity of the others tested. I expected they'd be at least close to what smaller companies offer.
The ThinkerNet does not reflect the views of TechWeb. The ThinkerNet is an informal means of communication to members and visitors of the Internet Evolution site. Individual authors are chosen by Internet Evolution to blog. Neither Internet Evolution nor TechWeb assume responsibility for comments, claims, or opinions made by authors and ThinkerNet bloggers. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
Ever since Google (Nasdaq: GOOG) hinted at an online music service similar to iTunes at the Google I/O conference in May, stories have been surfacing about it.
In what now seems like the ancient history of the technology industry, Sun Microsystems Inc. co-founder Scott McNealy talked about a future with “application dial tone.” Virtualization and cloud computing are getting us closer to that today, but there are still some major obstacles -- and many of them aren't technical.
The economy for the past several years has been brutal for companies across the board in the US, and the process improvement sector has been hit hard. Thousands of skilled process improvement professionals have found themselves unemployed. This has created several groups within the industry:
Last week, TechCrunch broke a rumor that Cisco Systems Inc. (Nasdaq: CSCO) has made a bid for Skype Ltd. Never mind that as recently as Friday, neither Cisco nor Skype would discuss the matter. Not that these noncommittal “official comments” mean much. The rumor was in the breeze, and that set people speculating about what this might mean for both companies.
Getting to Work on Smart Work: How IT Is Transforming the Implementation of the 'Internet of Things' Organizations in all industry sectors are becoming more instrumented, interconnected, and intelligent -- and that's changing the way they approach virtually every facet of their operations. It's up to IT to help organizations adopt a "Three I's" approach that leverages the emerging Internet of Things and enables them to work smarter. READ THIS eBOOK
your weekly update of news, analysis, and
opinion from Internet Evolution - FREE! REGISTER HERE
Wanted! Site Moderators Internet Evolution is looking for a handful of readers to help moderate the message boards on our site – as well as engaging in high-IQ conversation with the industry mavens on our thinkerNet blogosphere. The job comes with various perks, bags of kudos, and GIANT bragging rights. Interested?
To save this item to your list of favorite Internet Evolution content so you can find it later in your Profile page, click the "Save It" button next to the item.
High on the list of desired improvements from the mobile industry are: shared digital storage for the Internet; phone capability across borders; reduced electro-magnetic radiation; and rewards-based service plans.
Because 25% to 45% of broadband cost is due to sales and marketing, we could reduce our broadband prices by eliminating advertising and promotional spending by providers.
The next edition of one of the greatest English language reference books, the "Oxford English Dictionary," might not be published in paper. Bibliophiles might mourn, but should they?
RIM is giving in to demands by India to snoop on encrypted BlackBerry data. It's time to develop cheap or free encryption software for BlackBerrys and other cellular phones.
Nielsen’s recent numbers on the increasing use of texting bode well for enterprise networks. Shunning the phone in favor of text messaging could mean reducing bandwidth.
Two studios have filed suit against an ad broker for placing ads to help monetize P2P sites suspected of copyright infringement. That's taking a dangerous step toward what might be a worthy goal.
Digg
Del.icio.us
Reddit
Email This
TWEET THIS
