Packet Handoff Context Interface for External Pipeline Offload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network devices face challenges in efficiently offloading certain packet processing tasks to external devices, such as FPGAs, while maintaining control over the packet processing flow and ensuring seamless integration with internal pipelines.
Innovation Solution
A network device with a packet processing pipeline and handoff circuitry that generates an output context for external devices, receives an input context reflecting the external device's processing, and resumes processing based on this context, allowing for partitioning of packet processing between the internal pipeline and external devices like FPGAs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If packet processing tasks are offloaded to external devices, then processing capability and functionality are improved, but control over packet processing flow and integration with internal pipelines deteriorate
Solution Approach 1:
The patent divides packet processing into distinct segments: initial processing by the internal pipeline, handoff to external devices for specialized tasks, and resumption by the internal pipeline. This segmentation allows each component to focus on specific tasks while maintaining overall system control.
Solution Approach 2:
The patent introduces context data structures as intermediaries between the internal pipeline and external devices. These contexts carry packet state information, processing instructions, and results, enabling seamless handoff and resumption without losing control or integration.
2Adaptability or versatility
If packet processing is diverted to external devices, then advanced functionalities are enabled, but processing flow control and state management become more complex
Solution Approach 1:
The patent performs preliminary actions by generating context data structures that contain all necessary packet state information and processing instructions before handoff. This preparation ensures that external devices can process packets independently while maintaining proper state management.
Solution Approach 2:
The patent implements feedback mechanisms where external devices return processed packet information and results to the internal pipeline through the context data structures. This feedback loop enables the internal pipeline to resume processing with updated state information.
3Productivity
If multiple lookup engines process packets in parallel, then processing throughput is improved, but coordination and state retention become more difficult
Solution Approach 1:
The patent creates copies of packet context data structures for each parallel processing path. Each lookup engine receives its own context copy, processes packets independently, and returns results without interfering with other engines, simplifying coordination while maintaining high throughput.
Data Source
AI summary
A network device includes a packet processing pipeline and handoff circuitry. The packet processing pipeline is to apply to a packet a sequence of commands, one of the commands being a handoff command that diverts processing of the packet to an external device. The handoff circuitry is to generate, in response to the handoff command, an output context indicative of a current processing state of the packet, to send the output context to the external device, to receive from the external device an input context that (i) reflects the processing applied to the packet by the external device and (ii) specifies subsequent processing of the packet by the packet processing pipeline, and to forward the input context to the packet processing pipeline.