Database Storage Cache Configuration via Block Size Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current database storage systems face inefficiencies due to the mismatch between cache sizes and block sizes of tablespaces, leading to diminished cache advantages, as caches either cannot read or store entire blocks or remain unused.
Innovation Solution
The method involves mapping tablespaces with specific block sizes to corresponding logical units and caches of equal size, ensuring that tablespaces with unequal block sizes are not mapped to the same logical units, thereby optimizing cache utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If tablespaces with different block sizes are mapped to the same logical unit and cache, then storage capacity is utilized, but cache efficiency deteriorates because cache size does not match block size
Solution Approach 1:
The patent segments the storage system by creating separate logical units for different tablespace block sizes. Each logical unit is dedicated to a specific block size (e.g., 4KB, 8KB, 16KB), preventing mixing of different block sizes on the same logical unit. This segmentation allows caches to be optimized for their specific block size without waste.
Solution Approach 2:
The patent applies local quality by matching cache sizes to specific block sizes at each logical unit level. Each logical unit has a cache sized appropriately for its designated block size (e.g., a 4KB cache for 4KB blocks, an 8KB cache for 8KB blocks). This ensures that each local storage component has the optimal characteristics for its specific function.
2Quantity of substance
If cache size is smaller than tablespace block size, then memory usage is reduced, but read/write performance deteriorates because entire blocks cannot be read or stored
Solution Approach 1:
The patent performs preliminary action by pre-configuring logical units and their associated caches with matching block sizes before data is stored. This ensures that when data is written to a logical unit, the cache is already sized appropriately to hold entire blocks, enabling full-block read/write operations without requiring dynamic adjustment or causing performance degradation.
3Quantity of substance
If cache size is larger than tablespace block size, then more data can be cached, but cache utilization deteriorates because portions of the cache remain unused
Solution Approach 1:
The patent applies parameter changes by dynamically selecting the appropriate cache size parameter based on the tablespace block size. Instead of using a fixed cache size for all logical units, the system adjusts the cache size parameter to match the block size parameter for each logical unit. This ensures optimal cache utilization by eliminating both under-utilization and over-allocation.
Data Source
AI summary
A system and method for configuring storage resources for database storage are disclosed. A method may include mapping at least one first tablespace having a first block size to at least one first logical unit. The method may also include mapping the at least one first tablespace and the at least one first logical unit to a first cache having a size equal to the first block size. In addition, the method may include mapping at least one second tablespace having a second block size to at least one second logical unit. The method may further include mapping the at least one second tablespace and the at least one second logical unit to a second cache having a size equal to the second block size.