Modular SDN Switch Blade Segmentation for Packet Forwarding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional SDN switches face inefficiencies in handling elephant subflows and switch partitioning, as they require extensive processing by powerful processors for deep packet inspection, leading to high costs and resource utilization inefficiencies.
Innovation Solution
The implementation of a network architecture that includes Network Flow Switch (NFX) circuits and a Network Flow Processor (NFP) circuit, where NFX circuits are less powerful and do not execute an SDN protocol stack, allowing them to handle and forward packets efficiently by consulting an SDN flow table maintained by the NFP circuit, enabling cut-through switching for subsequent packets of subflows without involving the NFP for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional SDN switches use powerful processors for deep packet inspection and SDN protocol stack execution, then flow table management and packet inspection accuracy are improved, but processing overhead, cost, and resource utilization efficiency deteriorate
Solution Approach 1:
The system segments the SDN switch into multiple blade modules, each containing specialized components. The flow processor handles only control plane functions (SDN protocol stack and flow table management), while data plane packet forwarding is handled by dedicated forwarding engines in the blade modules. This segmentation allows precise packet inspection by the flow processor for only necessary control packets, while data packets are forwarded efficiently without full processor involvement, resolving the contradiction between inspection accuracy and processing efficiency.
Solution Approach 2:
The patent introduces an intermediary mechanism where the flow processor maintains flow tables and makes forwarding decisions, but actual packet forwarding is performed by dedicated forwarding engines in blade modules. For data packets, only flow table lookup is performed by the forwarding engine without involving the full SDN protocol stack. This intermediary approach maintains accurate flow-based forwarding while eliminating unnecessary processing overhead, resolving the contradiction between inspection accuracy and processing efficiency.
2Adaptability or versatility
If SDN protocol stack is executed on every switch blade, then each blade can independently process SDN messages, but device complexity and resource utilization deteriorate
Solution Approach 1:
The patent extracts the SDN protocol stack execution function from individual blade modules and consolidates it into a dedicated flow processor. Blade modules retain only the necessary flow table lookup and packet forwarding capabilities. This extraction allows the SDN protocol stack to be maintained in a single location, reducing overall system complexity while the flow processor can still provide centralized control to all blades, maintaining adaptability without duplicating the full protocol stack on each blade.
Solution Approach 2:
The flow processor serves as a universal control unit for all blade modules, handling SDN protocol stack execution and flow table management for the entire system. Instead of each blade having its own SDN protocol stack (reducing complexity), the universal flow processor provides centralized control, allowing blades to focus on their specialized data forwarding functions. This multi-functional approach maintains system adaptability while reducing individual blade complexity.
3Reliability
If all packets are forwarded through the control plane for flow table lookup, then flow table compliance is ensured, but processing time and resource utilization deteriorate
Solution Approach 1:
The system performs preliminary action by pre-computing and storing flow-based forwarding rules in flow tables maintained by the flow processor. Once a flow is established and its forwarding rule is determined, subsequent packets matching this flow can be forwarded by simple flow table lookup without re-invoking the full SDN protocol stack. This preliminary computation ensures flow table compliance while dramatically reducing processing time for subsequent packets in the same flow.
Solution Approach 2:
The patent implements a mechanism where packets that match existing flow entries can skip the lengthy SDN protocol stack processing and proceed directly to forwarding based on pre-determined flow table rules. The system rushes through the forwarding process for matched packets by bypassing unnecessary control plane involvement, while still ensuring flow table compliance through the pre-established flow rules. This skipping mechanism significantly reduces packet processing time while maintaining reliability.
Data Source
AI summary
An integrated circuit includes ingress ethernet ports and egress ethernet ports. A second ingress ethernet port is configurable to operate in a selected one of a command mode and a data mode. The ingress ethernet port does not power up in the command mode and can only be put into the command mode as a result of a port modeset command being received onto an ingress ethernet port operating in the command mode. A first ingress ethernet port powers up in the command mode. In the command mode the first ingress ethernet port can receive and carry out a port modeset command. Receiving and carrying out of the port modeset command causes one of the ingress ethernet ports identified by the port modeset command to operate in the command mode. A flow table structure adapted to store flow entries is used to determine which egress ethernet port outputs a packet.