Encrypted Traffic Classification
Know your traffic. Optimize efficiency. Assure QoS and Add Value.
Identify what’s using your bandwidth and troubleshoot faster
Can you ensure good network performance and high quality of service when faced with encrypted traffic? Allot helps you overcome this challenge so you can accurately identify all network traffic, understand how your network is being used, and respond accordingly to optimize your services, enhance users’ experience and create attractive value added services.
Get Past Encryption to Enhance Your Service
Accurate traffic classification is essential for achieving visibility of your network so that you can make the best decisions about traffic management, and it forms the foundation of the value you get with Allot multiservice solutions. However, traffic encryption that is widely used by Internet services and data privacy applications poses a serious challenge to traffic visibility. Encryption provides users with security and privacy by concealing the data flow and preventing its identification, but this makes it more difficult for you to control critical business applications, prioritize applications for optimal performance, manage network congestion, limit shadow IT and the use of unauthorized applications, such as recreational apps on enterprise networks. Allot enables you to by-pass encryption so that you can best manage traffic and confidently assure network efficiency, quality of service and users’quality of experience.
Traffic Classification with DART Accuracy
Allot’s superior DPI Traffic Classification is based on Dynamic Actionable Recognition Technology (DART) that employs multiple inspection methods to identify traffic according to Layer-7 application, user, device, access, video and contextual attributes. Allot’s synergy of inspection methods results in highly granular and accurate recognition even at maximum speeds and peak loads.DART proactively learns and adapts to the changing tactics of services and applications that use encryption, by using:
- Pattern and numerical property analysis of packet contents.
- Heuristic analysis of behavior statistics from inspected transactions.
- Peer learning: analysis of peer system behavior to identify P2P seeders (multiple transmission of files) and popular peers
- Port, IP address, or range of IP addresses classification
- Machine learning based on statistical distribution patterns for over 1000 traffic attributes
- Automatic IP discovery through analysis of service hosts and subscriber records
- Predictive DPI (PDPI): Allot’s patented technology that enables the classification engine to learn from the patterns and behavior of traffic that it recognizes with 98% accuracy, and compare them to encrypted flows in order to improve identification.
Network Visibility, Security and Control in a single integrated platform