Asynchronous Filter - Block Skype and P2P seamlessly

The Asynchronous Filter is software based and asynchronous, which means it's not real time, as it analyzes the flows a few seconds after they're actually recorded. It's non-destructive, and adds no latency to the network. Unlike common firewalls, its presence cannot be detected. It can detect any protocol, regardless of their stealthiness.

It usually takes 2 to 3 sec to identify a protocol after the first packets of the given flow are captured on the wire. After detection, rules can be implemented to discard, shape or record the given flow.

This filter and blocking system is entirely software. The asynchronous filter can run on top of most operating systems and is easy to manage. You can test it on your Windows box now for example.

Learn more about how the asynchronous filter works to block Skype and P2P and how you can test it now on the product page.

Opensource software

IT managers want control over their network. Whether for governement-based ISPs or corporate organizations, things like monitoring, filtering, shaping or blocking application are a common requirement. It is becoming unacceptable to let unwanted flows use corporate networks without the consentment of the pipes owners. Proprietary applications like Skype are also an growing threat to network security and individual privacy. Crucial applications like VoIP should obviously rely on open standards.

Therefore, we offer to the community an open and efficient Skype and P2P filter. Blocking Skype is a requirement for any serious organization concerned about security and privacy. Blocking P2P is mandatory for ISPs who want to prevent copyrighted material from flowing through their pipes. P2P shaping is also a must for ISP willing to spare their peering bandwidth.

And we made this Skype and P2P blocker opensource, so that any organization can integrate it safely in its environment for free.

The innerworkings: on-the-fly datamining

It's traditionally difficult to block Skype or P2P, as its designers make it purposely stealthy. To detect a stealthy and encrypted protocol, we came with datamining techniques that allow universal traffic detection and separation. As statistics is the core competency at Lynanda, it was obvious for us to offer a solution to the community. Bayesian Networks are at the basis of this network flows classification system. As flows are recorded on the wire, they feed a bayesian network that classify them. Unlike in traditional networks filters, the packet payload is not looked at. Behavioral data is not studied either (it's not an IDS).

Universal Traffic segregation

What can the Asynchronous Filter do as well:

• Allow distinction between all Skype usages (Skype2Skype, SkypeOut, SkypeChat, Skype FileTransfer, etc.)
• Allow to detect Edonkey or Limewire, even in encrypted mode
  
• Future-proof, as it doesn't rely on application signature, ie the filter works with any application or version, whether Skype 2.5, 3.0 or 4.0
• Allow the IT staff to fully understand the skype filtering or blocking process, because it's opensource and achieves a good level of modularity