Hybrid Register Storage for Fast IC Verification Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Verifying the functionality of large volumes of registers in semiconductor integrated circuits is challenging due to the voluminous data associated with registers, which makes finding and propagating register information tedious and slow, especially when register information is changed in HDL.
Innovation Solution
A hybrid storage approach is employed, where structured information is stored in a local database optimized for key-value pairs and unstructured information is stored in a cloud computing environment, enabling rapid data retrieval and utilizing cloud-based features for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If all register information is stored locally in a single database, then data access speed is fast, but storage requirements and data management complexity increase significantly
Solution Approach 1:
The patent segments register information into two categories: structured information (addresses, names, sizes, access types) stored locally in a first database, and unstructured information (text descriptions, usage contexts) stored remotely in a second database. This segmentation allows the system to keep critical access data locally for fast retrieval while moving bulk descriptive data to remote storage, thereby reducing local storage requirements while maintaining fast access speeds for essential register information.
2Ease of operation
If all register information is stored locally in a single database, then data retrieval is simple, but verification and analysis capabilities are limited
Solution Approach 1:
The patent creates a universal data management system where the first database handles structured data storage and retrieval, while the second database provides enhanced verification, validation, and analysis capabilities for unstructured data. The system can operate in multiple modes: simple local retrieval for routine operations, and comprehensive remote verification for validation tasks. This multi-functional architecture allows the same system to serve both simple data access needs and complex verification requirements without compromising either capability.
3Reliability
If register information is propagated when HDL changes, then data accuracy is maintained, but propagation time and computational overhead increase
Solution Approach 1:
The patent applies local quality by storing structured register information (addresses, names, sizes, access types) locally in the first database, which can be quickly updated when HDL changes occur. Since structured data represents the core register definitions that change most frequently with HDL modifications, keeping this data locally enables rapid propagation of changes without the overhead of updating entire datasets. Remote unstructured data remains unchanged unless explicitly modified, reducing overall propagation time while maintaining data accuracy through the localized update mechanism.
Data Source
AI summary
Embodiments of the present disclosure include a method of storing information about registers in an integrated circuit. In one embodiment, registers from an integrated circuit have corresponding structured and unstructured information describing the registers. In one embodiment, validation software stores the structured information in one database and stores the unstructured information in another database. The structured information may be stored in a high-speed local database, and the unstructured information may be stored in a cloud computing environment database with access to cloud software services. For each register, structured information in the local database may be linked to structured information in the cloud database using a unique ID.