Region-Based Locking for 3D Spatial Data Concurrency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing locking strategies for fully-connected large-scale multi-dimensional spatial data, such as 3-D meshes, are inefficient and impractical due to reliance on tile boundaries, object-based structures, file-level locking, and row-level locking, which restrict concurrency and hinder processing on available hardware resources.
Innovation Solution
Implementing region-based locking techniques using bounding boxes to manage access and locks, allowing clients to access data based on the intersection of bounding boxes and their lock states, enabling exclusive or sharable locks to ensure data consistency and concurrency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tile-based locking strategies are used to maintain data consistency, then conflicting writes are prevented, but client concurrency is restricted and processing efficiency deteriorates
Solution Approach 1:
The patent segments the locking mechanism from the data structure by introducing region-based locks that operate independently of tile boundaries. Clients can lock arbitrary spatial regions defined by bounding boxes without being constrained by predefined tile structures, allowing finer-grained concurrent access while maintaining data consistency through region-based conflict detection.
Solution Approach 2:
The locking strategy transitions from static tile-based locks to dynamic region-based locks. Lock regions are determined at runtime based on client access patterns and bounding box intersections rather than fixed predetermined tiles, enabling adaptive concurrency optimization while preserving data integrity.
2Productivity
If object-based locking strategies are used to allow concurrency on well-separated objects, then processing efficiency improves, but the approach becomes impractical when data lacks extensive structure
Solution Approach 1:
The region-based locking mechanism serves multiple purposes: it enables concurrent access for structured data with known objects, provides efficient locking for unstructured data using bounding boxes, and maintains compatibility with various data formats including meshes, point clouds, and volumetric data. This universal approach eliminates the need for elaborate data structuring.
Solution Approach 2:
The patent changes the fundamental parameter for locking from object identity to spatial region coordinates. By representing lock targets as bounding boxes defined by minimum and maximum coordinates rather than specific object references, the system becomes adaptable to both structured and unstructured spatial data without requiring pre-existing object hierarchies.
3Reliability
If file-level locking is used to preserve atomicity, then data integrity is maintained, but performance deteriorates and task independence is lost
Solution Approach 1:
The patent segments the file-level lock into multiple fine-grained region-level locks. Instead of locking entire files, the system locks only the specific spatial regions that clients need to access, allowing other clients to simultaneously access different regions of the same file. This segmentation maintains atomicity within regions while enabling parallel processing across regions.
Solution Approach 2:
The locking strategy applies local quality control by enforcing atomicity only where necessary - at the region level rather than file level. Each region maintains its own lock state, allowing different quality levels of protection for different parts of the data based on actual access patterns and conflict potential, optimizing both integrity and performance.
4Productivity
If row-level locking in RDBMS is used to allow concurrent access, then concurrency improves, but the approach becomes impractical when data volume exceeds hardware resources
Solution Approach 1:
The patent extracts the locking mechanism from the database system and implements it at the application level using spatial bounding boxes. This removes the need for row-level database locking and associated overhead, reducing hardware resource requirements while maintaining high concurrency through efficient spatial intersection checks.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
In an example embodiment, techniques are provided for locking a region of fully-connected large-scale multi-dimensional spatial data (e.g., a large-scale 3-D mesh) defined by a bounding box. A region is associated with a lock state (e.g., exclusive or sharable). Clients may access the fully-connected large-scale multi-dimensional spatial data based on a comparison of the bounding box of the requested spatial data to the bounding boxes of other client's locks and their lock state.