Pensando Tree LPM Lookup for Scalable Network Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current network systems face challenges in providing flexible network interfaces to adapt to changes in device structure, protocols, and protocols, and they struggle with high CPU usage due to increased network complexity, especially with the introduction of overlay tunneling protocols and virtual switching, leading to inefficiencies in CPU resource allocation and memory usage.

Innovation Solution

A programmable IO device employing a Pensando Tree architecture with multiple cascading stages, each performing arithmetical operations, and a match processing unit, which sorts range-based elements, generates interval tables, and compresses data into cache-lines to optimize memory access and reduce memory requirements, allowing for efficient longest prefix match lookups without storing memory addresses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If network interfaces are made flexible to adapt to changes in device structure and protocols, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improvenetwork interface flexibilityVSAvoidcontrol structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a programmable IO device where the control structure can be dynamically reconfigured through software loading. The device transitions from a fixed control structure to a dynamic one where control logic can be updated without hardware changes, resolving the contradiction between adaptability and complexity by making the system configurable rather than static

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The programmable IO device serves multiple functions including network interface control, packet processing, and protocol adaptation within a single unified architecture. This universal device replaces multiple specialized fixed-function devices, improving adaptability while managing complexity through consolidation

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If network processing functions are performed in software to provide flexibility, then adaptability is improved, but CPU usage increases

Engineering Contradiction:

Improvenetwork processing flexibilityVSAvoidCPU resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent introduces a programmable IO device as an intermediary between the CPU and network interface. This mediator handles network processing tasks that would otherwise require full CPU intervention, reducing CPU usage while maintaining flexibility through programmable control logic in the intermediate device

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system segments network processing functions between the programmable IO device and the CPU. The IO device handles lower-level packet processing and control, while the CPU focuses on higher-level application tasks, dividing the workload to reduce overall CPU consumption while maintaining system flexibility

Inventive Principle:

Principle #1Segmentation

3Use of energy by moving object

If fixed-function offload technologies are used to reduce CPU load, then CPU usage is reduced, but adaptability deteriorates

Engineering Contradiction:

ImproveCPU resource availabilityVSAvoidnetwork function fungibility

Core Design Contradiction:

Use of energy by moving objectVSAdaptability or versatility

Solution Approach 1:

The programmable IO device uses dynamic configuration where control logic can be reloaded and reprogrammed to perform different network functions. This dynamic approach allows the device to adapt to changing requirements while still offloading processing from the CPU, unlike static fixed-function devices

Inventive Principle:

Principle #15Dynamics

4Ease of manufacture

If traditional LPM lookup methods are used in distributed systems, then implementation is simplified, but scalability deteriorates

Engineering Contradiction:

Improveimplementation simplicityVSAvoidsystem scalability

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent segments the LPM lookup operation into distributed components where each node maintains a portion of the routing information. This segmentation allows the system to scale by adding more nodes, each handling local lookups, while maintaining implementation simplicity through standardized node behavior and modular architecture

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20220060419A1Systems for building data structures with highly scalable algorithms for a distributed LPM implementation

Publication Date: 2022.02.24 PENSANDO SYSTEMS INC

AI summary

Described are programmable IO devices configured to perform operations. These operations comprise: determining a set of range-based elements for a network; sorting the set of range-based elements according to a global order among the range-based elements; generating an interval table from the sorted range-based elements; generating an interval binary search tree from the interval table; propagating data stored in subtrees of interior stages of the interval binary search tree to subtrees of a last stage of the interval binary search tree such that the interior stages do not comprise data; converting the interval binary search tree to a Pensando Tree; compressing multiple levels of the Pensando Tree into cache-lines; and assembling the cache-lines in the memory unit such that each stage can compute an address of a next-cache line to be fetched by a next stage.