MIPS Analysis Tool for 5G Data Stack Core Configuration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional MIPS calculation tools are oversimplified and fail to accurately reflect the actual field application, making it difficult to explore data stack design options for 5G data communication systems.

Innovation Solution

A MIPS analysis tool that performs a Monte Carlo simulation based on user-specified use cases, processor specifications, and traffic models to determine the recommended configuration of processor cores for 5G data stacks, using a user interface to input requirements and generate simulation results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional MIPS calculation tools are used, then the calculation process is simple, but the accuracy of reflecting actual field application is poor

Engineering Contradiction:
Improveaccuracy of MIPS calculationVSAvoidcomplexity of analysis tool
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining traffic models with multiple packet sizes and their allocation distributions before performing MIPS calculations. This allows the tool to simulate real-world traffic patterns in advance, improving the accuracy of MIPS measurements without requiring complex real-time monitoring during actual calculations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements dynamics by enabling the traffic model to be dynamically configured with different packet size distributions and allocation percentages. This allows the MIPS analysis tool to adapt to various actual field application scenarios, making the measurement both accurate and flexible without requiring a completely new tool for each scenario.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If detailed traffic models with multiple packet sizes are used in Monte Carlo simulation, then the simulation accuracy improves, but the computational complexity increases

Engineering Contradiction:
Improvesimulation result accuracyVSAvoidcomputational resources required
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The patent applies partial action by implementing a multi-stage simulation approach where critical path functions are simulated in detail while non-critical functions use simplified models. This allows the Monte Carlo simulation to focus computational resources on the most impactful components, achieving high accuracy for MIPS measurement without requiring excessive computational power across the entire system.

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If comprehensive processor specifications and user-specified functions are analyzed, then the recommended configuration accuracy improves, but the analysis time increases

Engineering Contradiction:
Improveconfiguration recommendation accuracyVSAvoidanalysis time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent applies segmentation by dividing the processor analysis into distinct functional components (e.g., packet processing functions, protocol handling, data plane operations). Each function is analyzed separately with its own performance characteristics, allowing the tool to provide accurate configuration recommendations while reducing overall analysis time through modular processing of each function segment.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20230124083A1Data stack MIPS analysis tool for data plane
Publication Date: 2023.04.20 NOKIA TECHNOLOGIES OY
  • US20230124083A1 patent drawing
  • US20230124083A1 patent drawing
  • US20230124083A1 patent drawing

AI summary

Apparatus and methods for performing a Million Instructions per Second (MIPS) analysis for a data stack of a user equipment (UE) are disclosed. The method includes (i) receiving an input for a Monte Carlo simulation, the input including a requirement for one or more use cases, a processor specification, and a user-specified function; (ii) determining a traffic model, a number of packets to be run for each use case, and a seed value for the Monte Carlo simulation; (iii) performing the Monte Carlo simulation based on the input and the traffic model to generate a simulation result; and (iv) determining a recommended configuration of processor cores for the data stack based on the simulation result.