Hierarchical Instance Name Compression for Circuit Simulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional hierarchical systems for circuit simulation face challenges in efficiently organizing and retrieving names of instances, leading to increased data usage and performance issues as the number of components in integrated circuits grows, especially at nanometer scales.

Innovation Solution

A computer-implemented method and system that creates an instance name data structure using a map of indexes and an array of offsets to access naming information in a hierarchical system, associating instance names with unique integers arranged sequentially, allowing for efficient data compression and improved simulation performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If conventional hierarchical systems use detailed naming for each instance in a hierarchical circuit system, then the system can accurately represent and retrieve instance information, but the data size becomes prohibitively large (20 GB for systems with millions of components)

Engineering Contradiction:
Improveinstance naming information completenessVSAvoiddata size
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent segments the hierarchical instance naming system into multiple levels (root level, intermediate levels, and leaf level). Each level maintains only the necessary naming information for its scope, rather than storing complete instance names at every level. This segmentation reduces redundant data storage while preserving the ability to accurately represent and retrieve instance information through hierarchical path construction.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and stores only the essential naming components at each hierarchical level, rather than storing complete instance names. By taking out only the necessary path information needed to reconstruct full instance names, the system significantly reduces data size while maintaining information completeness for simulation operations.

Inventive Principle:
Principle #2Taking out (Extraction)

2Ease of operation

If the system stores complete instance names for all components in hierarchical levels, then instance retrieval is straightforward, but the memory requirements become unmanageable for circuits with millions of devices

Engineering Contradiction:
Improveinstance name retrievalVSAvoidmemory requirements
Core Design Contradiction:
Ease of operationVSQuantity of substance

Solution Approach 1:

The patent divides the naming system into hierarchical segments where each level stores condensed path information. This segmentation enables efficient retrieval by allowing the system to reconstruct complete instance names on-demand from the segmented path data, rather than storing and searching through complete names for all instances.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from storing instance names in a traditional flat or single-dimension structure to a multi-dimensional hierarchical structure. By organizing naming information across multiple hierarchical dimensions (levels), the system enables efficient retrieval operations while dramatically reducing the total data volume required to represent millions of circuit components.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Measurement precision

If detailed circuit simulation is performed on systems with millions of components, then simulation accuracy is maintained, but simulation performance deteriorates due to the large amount of data to be processed

Engineering Contradiction:
Improvesimulation accuracyVSAvoidsimulation performance
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the circuit system into hierarchical levels with condensed naming representations. This segmentation reduces the data processing burden during simulation while maintaining the ability to accurately reference and simulate individual components, thereby improving simulation performance without sacrificing accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a compressed hierarchical naming structure that serves as an efficient copy or representation of the complete instance naming system. This compressed structure enables faster data processing and simulation operations while the full instance names can be reconstructed on-demand when detailed information is needed for accurate simulation.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS8819086B2Naming methodologies for a hierarchical system
Publication Date: 2014.08.26 PRIMARIUS TECH CO LTD
  • US8819086B2 patent drawing
  • US8819086B2 patent drawing
  • US8819086B2 patent drawing

AI summary

Methods and systems are disclosed for naming methodologies for a hierarchical system. In one embodiment, a computer implemented method of organizing instance names in a hierarchical system includes receiving a description of a hierarchical system that includes plurality of instances arranged in different branches in a plurality of hierarchical levels in a physical data structure, creating an instance name data structure configured to describe the corresponding instances in the hierarchical system, where the instance name data structure comprises a map of indexes and a corresponding array of offsets configured to access naming information in a subsequent level, and associating names of instances in the hierarchical system to a corresponding set of unique integers which are arranged in a sequential manner.