The TCP Acceleration Performance Enhancing Proxy (PEP) is an optional feature that can be enabled on the Wan Accelerator to significantly enhance performance when sending TCP traffic over suboptimal network conditions or to improve TCP performance of slow TCP stack implementations.
Note: The PEP TCP Accelerator is available in v.4.
Limitations of Standard TCP
Most applications such as Email, HTTP, FTP and TCP based SSL VPN, use TCP. Standard TCP congestion control estimates the safe transmission rate in the network and has throughput fall-back mechanisms in case of suspected congestion. TCP performance can be limited by the host TCP stack implementation, processing speeds of routers, bandwidth available in intermediate links, packet loss, round trip delay, changes in round trip delay and other factors.
Standard TCP also has two major drawbacks for the long delay environment:
• Long delays in the network leads to suboptimal window size calculations.
• TCP Slow start mechanism avoids burst on low latency links and doubles throughput every round trip time, which leads to slow TCP throughput over long delay links.
Overview of the TCP Accelerator
The TCP Accelerator is a full transparent performance enhancing proxy that can be enabled to transparently optimize TCP sessions on the fly. The TCP accelerator utilizes the following mechanisms to improve TCP Performance:
HighSpeed TCP – Optimize TCP performance with HighSpeed TCP for high-speed high-delay networks.
Window Scaling – The TCP Window Scale option is used to set the optimal window sizes in order to avoid the throughput limitation of a suboptimal TCP window size.
Selective Acknowledgement – When packets are lost and unrecoverable through the packet loss recovery feature, the sender is notified to only retransmit the missing packets.
Local Congestion and Flow Control – TCP Accelerator splits an end-to-end TCP connection transparently and operates each link separately. As the bandwidth will typically be different between the sections, the TCP Accelerator use different HighSpeed TCP congestion and flow control metrics for each.
Local Acknowledgements – The TCP session acknowledgements are sent locally between the proxy and server, as well as the proxy and client to avoid idle time waiting for acknowledgements.
The result of these features working together is faster file transfers for TCP-based applications. Where TCP throughput is suboptimal due to the limitations of standard TCP or network conditions, TCP speed improvements with the TCP Accelerator enhancements are expected to range from 2-10X. The speed improvement depends on the severity of conditions that hinder performance. Total speed improvement is also dependent on the optimization ratio of other acceleration features like Stream Compression, Deduplication and Packet Loss Recovery.
The following diagram shows the TCP Accelerator deployment.
Note: Some conditions exist where the TCP Accelerator will not enhance performance:
• Non-TCP based protocols e.g. UDP based video streaming or audio streaming
• TCP Acceleration can be diminished in optimal network conditions e.g. high speed low latency environments.