Key Benefits:
•  Eliminates 65% to 98% of TCP packets required to transfer data
•  Accelerate encrypted (SSL) traffic like HTTPS
•  Optimizes both low-bandwidth and high-bandwidth connections

Transport Streamlining improves the performance of TCP-based applications by improving the way transport protocols behave on WANs, reducing the number of TCP packets required to transfer data by 65% to 98%. Transport Streamlining overcomes the limitations of TCP by adapting transmission characteristics such as window scale, loss handling, congestion notification, and more.

Transport Streamlining now supports the acceleration of encrypted SSL traffic (like HTTPS), which means that even your secure business applications can benefit from Riverbed’s award-winning application acceleration.

For high-bandwidth WAN links (also known as “Long Fat Networks”, or “LFNs”) components of Transport Streamlining known as high speed TCP (HS-TCP) and Max-Speed TCP (MX-TCP) may be activated which enables greater bandwidth utilization, providing the capability to “fill the pipe” more effectively. MX-TCP also helps dealing with lossy network connections.

RiOS is designed to adapt to network conditions on-the-fly, responding to events such as congestion or packet loss without giving up the reliability and scalability that make TCP the de facto standard. With Transport Streamlining, enterprises can be assured that their networks are optimally transferring data.

Unlike other products which may implement one or two features designed to optimize TCP, RiOS implements both patented and industry-accepted features in order to fully maximize the power of TCP:

• Virtual Window Expansion
• Window Scaling
• Delayed and Selective Acknowledgments (RFC 2018)
• Explicit Congestion Notification (RFC 3168)
• Limited and Fast Re-Transmits (RFC 3042 and RFC 2582)
• Adaptive Initial Congestion Windows (RFC 3390)
• Slow Start with Congestion Avoidance (RFC 2581)
• Bandwidth Delay Control
• Download the RiOS Technical Overview to learn more about these features.

MX-TCP vs. HS-TCP

Most terrestrial WANs have a very low natural packet loss rate, normally well under 0.1%, and often under 0.01%. Some enterprises may be dealing with lossy or dirty circuits however, so it is imperative that any WAN optimization solution knows how to deal with both clean and lossy connections.

On clean long fat network (LFN) circuits, features like HS-TCP enable higher bandwidth utilization because they ramp up faster and back off more slowly in the face of any congestion or loss. HS-TCP behaves like regular TCP when dealing with other traffic on the same connection it will back off so all senders can transmit on the link.

On lossy LFN circuits, or on clean LFN circuits where an administrator can guarantee a certain amount of bandwidth, MX-TCP is a better solution. MX-TCP will use 100% of an allocated amount of bandwidth until a transfer has been completed. It will not back off in the face of packet loss, it merely resends any lost packets.