Variable-Size Memory Block Allocation for Database Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory allocation methods in computer systems are inefficient as they cannot adapt to dynamic changes in storage space requirements, leading to suboptimal utilization of storage resources and increased memory fragmentation, which reduces computer performance.
Innovation Solution
A storage space allocation method that dynamically manages virtual and physical address blocks by monitoring addition and deletion operations in a database, allowing for real-time adjustment of storage space allocation and release, and merging adjacent free blocks to reduce fragmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a fixed-size paging management method is used for memory allocation, then the memory allocation process is simple and manageable, but the storage resource utilization rate decreases due to inability to use free memory smaller than fixed space and increased memory fragmentation
Solution Approach 1:
The patent segments the memory pool into multiple variable-size memory blocks instead of using fixed-size pages. Each memory block's size is determined by actual allocation needs, allowing fine-grained memory management. This segmentation enables the system to allocate memory efficiently without wasting space, directly resolving the contradiction between allocation simplicity and resource utilization.
Solution Approach 2:
The patent implements dynamic memory block size adjustment based on actual allocation requirements. The memory allocator can create memory blocks of different sizes rather than being constrained to fixed page sizes. This dynamic approach allows the system to adapt to varying memory needs while maintaining simple allocation procedures, thereby improving storage resource utilization without complicating the allocation process.
2Ease of manufacture
If a fixed-size paging management method is used for memory allocation, then the memory allocation process is straightforward, but computer performance decreases due to memory fragmentation and reduced use efficiency
Solution Approach 1:
By segmenting memory into variable-size blocks rather than fixed pages, the system eliminates internal fragmentation while keeping the allocation process straightforward. The patent maintains simple allocation logic while improving performance through efficient memory utilization, directly addressing the contradiction between process simplicity and computer performance.
Solution Approach 2:
The patent changes the fixed size parameter of memory blocks to a variable parameter based on actual needs. This parameter change allows the system to optimize memory usage patterns, reducing fragmentation and improving access efficiency, thereby enhancing computer performance without complicating the allocation process.
3Device complexity
If traditional memory allocation methods are used, then the implementation is simple, but memory fragmentation increases leading to reduced memory use efficiency and computer performance
Solution Approach 1:
The patent implements dynamic memory block size management where blocks can be allocated, freed, and merged based on actual usage patterns. This dynamic approach prevents fragmentation accumulation while maintaining relatively simple allocation logic, resolving the contradiction between implementation simplicity and memory use efficiency.
Solution Approach 2:
The patent merges adjacent free memory blocks when they become available, creating larger contiguous free spaces. This merging operation reduces fragmentation without adding significant complexity to the allocation system, directly addressing the contradiction between device complexity and memory use efficiency.
Data Source
AI summary
The present application provides a storage space allocation method. The method includes allocating a first virtual address block list based on an initial storage space requirement of each table object in a database. A first correlation and a second correlation are obtained. Based on the first correlation and the second correlation, a first virtual address and a first physical address which correspond to first data in the table object and a second physical address and a second virtual address of second data are obtained after performing an addition/deletion operation on the first data. Once a storage status of the second physical address and a storage status of a second physical block are updated, a storage status of the second virtual address and a storage status of a second virtual block are synchronously updated.