Guest blogger, Aron Rosenberg, Founder & CTO of SightSpeed, responds to Om Malik's critical and provocative question that literally is around the blogosphere and media world-wide -- i.e., "Does Skype’s outage indicate a fatal flaw in using Peer-to-Peer?"
Bottom line -- according to Aron (his full analysis/opinion is below), all Peer-to-Peer models are not created equal. Skype's Peer-to-Peer model suffers from flaws not present in other Peer-to-Peer models (so it is Skype-specific vulnerabilities that are exposed by its ongoing outage, rather than vulnerabilities to Peer-to-Peer in general).
How is Skype different from other Peer-to-Peer models?
According to Aron, like its predecessor Kazaa, Skype uses a different type of Peer-To-Peer network than most companies. Skype uses a system called SuperNodes. A SuperNode Peer-to-Peer system is one in which you rely on your customers rather than your own servers to handle the majority of your traffic. SuperNodes are just normal computers which get promoted by the Skype software to serve as the traffic cops for their entire network. In theory this is a good idea, but the problem happens if your network starts to destabilize. Skype, as a company, has no physical or programmatic control over the most vital piece of its product. Skype instead is at the mercy of and vulnerable to the people who unknowingly run the SuperNodes.
This of course exposes vulnerabilities to any business based on such a system -- systems that, in effect, are not within the company's control.
According to Aron, another flaw with SuperNode models concerns system recovery after a crash. Because Skype lost its SuperNodes in the initial crash, its network can only recover as fast as new SuperNodes can be identified.
Other companies such as SightSpeed that use Peer-to-Peer do not suffer from these flaws that proved to be fatal in this case, since these companies manage and handle all the core functionality themselves. Standards based telephony protocols such as SIP (which SightSpeed uses) were designed from the outset to be fault tolerant.
Thanks Aron -- appreciate your thoughts on this important topic.
(And, if you would like to check out SightSpeed as an alternative, SightSpeed now has a special offer for Skype users and others to check out its full premium services, SightSpeed PRO, FREE for 30 days -- SightSpeed's regular service is always free.)
Friday, August 17, 2007
Subscribe to:
Post Comments (Atom)
Posts
4 comments:
That's a little big disingenuous. Both approaches have their flaws, Sightspeed's would be that you have all your eggs in one basket, and if their central authentication and management servers go down, or DNS gets screwy, or whatever, you lose everything.
I don't think there's any such thing as a truly fault-tolerant VOIP design, just different approaches that have different trade-offs. We're seeing one BIG trade-off with the SuperNode concept right now.
Supernode architecture is rock solid, the Skype bug which took down the network (apparently in the code since day 1, but dormant according to Skype) does expose its Achilles heal. Cascading supernode failure is the one golden egg you want to avoid at all costs, as you really have no backup when you rely on your own network to update your nodes.
As the previous commenter mentioned no system is 100% (although one could argue the PSTN is)
Sightspeed is really good btw: 3 quick things that would be nice to haves. Call transfer, Voicemail to Email, and a contact importer (grab outlook/gmail addies etc) if you had a contact importer today I would have invited my entire address book - but since Skype is pretty much back up you missed the boat a bit on that one.
And, please explain exactly what parts of SIP are 'designed from the outset to be fault tolerant'. SIP is a way of setting up and tearing down calls on top of an existing 'reliable' transport network. In itself it is not fault tolerant in any way.
Also, how can sightspped claim to be Peer to Peer and still have central severs. The two terms are mutually exclusive.
SIP uses DNS SRV records which have built-in load balanceing and fault-tolerance for servers.
SightSpeed uses central servers for the signaling and authentication protocol and Peer-to-Peer for the actual video and audio traffic. Skype uses Peer-to-Peer for both signaling, authentication and video+audio data.
Post a Comment