Multi-Dimensional Cube Access Path Protocol for Distributed Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In shared nothing architecture distributed computing systems, processing and querying multi-dimensional data is challenging due to large data sizes, storage complexity, and computational expenses, particularly when dealing with multi-dimensional cubes and queries that require aggregation and partitioning.
Innovation Solution
The system employs Expensive Dimension Level Materialization (EDLM), Hash Partitioning of Materialized Dimension Levels (HPMDL), Orthogonal Dimension Level Partitioning (ODLP), and Family Operation Resolution using Metadata (FORM) to pre-materialize expensive dimension levels, reduce computational costs, and optimize data storage and retrieval by using metadata to manage hierarchical and partitioned data structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multi-dimensional data is stored and queried in shared nothing architecture, then scalability is improved, but storage complexity and query processing difficulty increase
Solution Approach 1:
The patent segments the multi-dimensional cube into partitions distributed across multiple nodes, with each partition representing a subset of data. This segmentation enables scalability by allowing the system to handle large datasets across multiple nodes while reducing storage complexity at each individual node through distribution.
Solution Approach 2:
The patent introduces a metadata dimension that characterizes access paths to data within the multi-dimensional cube. This additional dimension organizes data based on query patterns and access requirements, transforming the storage structure to reduce complexity while maintaining scalability.
2Loss of time
If multi-dimensional cubes are materialized to improve query performance, then query time is reduced, but computational expense and storage requirements increase
Solution Approach 1:
The patent pre-materializes expensive dimension levels by computing and storing aggregated data for frequently accessed dimensions before queries are executed. This preliminary computation reduces query time significantly while the selective materialization approach limits computational expense to only the most necessary dimensions.
Solution Approach 2:
The patent applies different materialization strategies to different dimension levels based on their accessibility and computational cost. Expensive dimension levels that are frequently accessed are materialized, while less critical dimensions remain unmaterialized, optimizing the balance between query performance and computational resource consumption.
3Productivity
If data is partitioned to enable distributed processing, then processing efficiency is improved, but access path complexity increases
Solution Approach 1:
The patent introduces metadata as an intermediary structure that characterizes access paths to data within the multi-dimensional cube. This metadata layer acts as a mediator between queries and the distributed partitions, translating query requirements into simplified access paths that navigate the partitioned data structure without increasing operational complexity.
Data Source
AI summary
A computer readable storage medium includes executable instructions to define an interface to a multi-dimensional cube divided into partitions distributed across nodes supporting a shared nothing multi-dimensional distributed database. The interface includes a metadata module characterizing access paths to data within the multi-dimensional cube and an access path module to convert a data request into an access path protocol to retrieve data from the multi-dimensional cube to satisfy the data request.