Null-Elimination Data Slab Compression for Parallel Query Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems face limitations in processing speed due to hardware constraints, data storage methods, and restricted co-processing options, leading to inefficiencies in data handling and query execution.
Innovation Solution
A parallelized database system architecture that includes a parallelized data input, store, retrieve, and process sub-systems, along with a query and response system, utilizing a distributed network of computing devices and nodes with independent processing core resources to optimize data storage and query execution through techniques like null elimination and data slab compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional database storage and processing methods are used, then system simplicity is maintained, but processing speed and efficiency are limited due to hardware constraints and sequential execution
Solution Approach 1:
The database system is divided into multiple independent nodes, each capable of autonomous processing. Data is segmented into distributed partitions across these nodes, enabling parallel query execution and eliminating single-point bottlenecks. This segmentation transforms the system from a monolithic sequential processor into a distributed parallel processing architecture.
Solution Approach 2:
The system transitions from traditional single-node vertical processing to a multi-dimensional distributed architecture. Queries can be executed across multiple nodes simultaneously, adding a spatial dimension to processing throughput. The parallel execution framework operates in an additional computational dimension, multiplying overall system capacity without linearly increasing individual node complexity.
2Productivity
If data is stored in traditional formats with null values, then data completeness is maintained, but storage efficiency and processing speed are reduced
Solution Approach 1:
Null values are extracted and eliminated from the data storage structure. The system uses null elimination techniques where missing values are not stored as explicit null markers but are handled through compact encoding schemes. This extraction of null value overhead reduces storage volume and eliminates the processing burden of tracking and managing null indicators throughout the data pipeline.
Solution Approach 2:
The data storage format undergoes parameter changes from traditional row-based storage with explicit null fields to columnar storage with null elimination. By changing the fundamental storage parameters to use variable-length encoding and only storing non-null values, the system achieves superior compression ratios while maintaining query performance through vectorized processing of compact data structures.
3Productivity
If lock-based concurrency control is used, then data consistency is ensured, but execution speed is reduced due to locking overhead and contention
Solution Approach 1:
The mechanical lock-based concurrency control system is replaced with a software-based optimistic concurrency control mechanism. Instead of using physical or logical locks that block execution, the system uses version vectors and conflict detection algorithms that allow parallel execution. Transactions proceed without blocking, and consistency is ensured through conflict resolution at commit time rather than through preventive locking.
Solution Approach 2:
The system enables continuous query execution without interruption from lock acquisition and release cycles. Parallel query processors can continuously scan and process data streams without being blocked by transaction locks. This continuity is achieved through architecture that separates data access from transaction commit, allowing read operations to proceed continuously while write operations are coordinated through less intrusive mechanisms.
Data Source
AI summary
A data input sub-system of a parallelized database system includes processing core resources operable to obtain divisions of data slabs of a dataset and compress the divisions of data slabs using a null elimination compression scheme to produce divisions of compressed data slabs, A first data slab of a first division of data slabs of the divisions of data slabs is compressed using the null elimination compression scheme to produce a first compressed data slab. The first compressed data slab includes first compressed data and first compression information. The processing core resources are further operable to store a respective division of compressed data slabs of the divisions of compressed data slabs.


