In the 6WINDGate software architecture, the control plane and data plane are separate.
Within the data plane, the 6WINDGate fast path runs isolated from the Linux operating system, on a dedicated set of processor cores. The fast path protocols process the majority of network packets, without incurring any of the Linux overheads that degrade overall performance. The fast path implements a run-to-completion model, whereby all cores run the same software and can be allocated as required, according to the necessary level of packet processing or Linux application performance. Please click here to learn more about how 6WINDGate provides best-in-class packet processing performance.
Only those rare packets that require complex processing are forwarded to Linux, which performs the necessary management, signaling and control functions. Packet processing information that is configured or learned (through control plane protocols) in Linux is automatically and continuously synchronized with the fast path, so that the presence of the fast path is completely transparent to Linux and its applications. Please click here to learn more.
This optimized architecture is ideal for high-performance networking systems, in which the vast majority of the processing workload is data plane packet processing, rather than control plane and signaling functions.
By using a Linux userspace model for the fast path, the system can be reconfigured dynamically as traffic patterns change, in order to optimally share the CPU resources allocated to the control plane and the fast path.
The 6WINDGate architecture enables major performance and utilization improvements for network virtualization in SDN applications.
To learn more about this topic, please click here to visit our FAQ section.