Server Managed Cache Segmentation for Write Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As storage capacity increases, the processing load for finding writable areas in cache memories within storage subsystems grows, making it difficult to maintain high access performance.
Innovation Solution
A computer system with a server that manages cache area usage in the storage subsystem, allowing data to be directly written to writable areas without searching, thereby reducing the processing load during write processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the storage capacity of the storage subsystem is increased, then the cache memory capacity is increased, but the processing load for searching writable areas increases
Solution Approach 1:
The cache memory is divided into multiple regions with different access characteristics. The first region allows normal processor access, while the second region is optimized for direct write operations. This segmentation enables the system to handle large cache capacities without proportionally increasing the search load, as the second region can be directly written to without extensive searching.
Solution Approach 2:
The system pre-allocates and manages writable areas in the cache memory before write operations occur. The server judgment mechanism determines in advance whether writable areas exist and prepares them for direct writing, eliminating the need for real-time search operations during actual write processes. This preliminary preparation significantly reduces processing load during data writes.
2Quantity of substance
If the cache memory capacity is increased, then more data can be stored, but the time to search for writable areas increases
Solution Approach 1:
The cache memory is divided into multiple regions with different access characteristics. The first region allows normal processor access, while the second region is optimized for direct write operations. This segmentation enables the system to handle large cache capacities without proportionally increasing the search load, as the second region can be directly written to without extensive searching.
Solution Approach 2:
The system pre-allocates and manages writable areas in the cache memory before write operations occur. The server judgment mechanism determines in advance whether writable areas exist and prepares them for direct writing, eliminating the need for real-time search operations during actual write processes. This preliminary preparation significantly reduces processing load during data writes.
3Adaptability or versatility
If the storage subsystem processes write requests independently, then it maintains autonomy, but the access performance deteriorates with increased storage capacity
Solution Approach 1:
The system merges the write processing functions between the storage subsystem and the server. The server participates in judging whether writable areas exist and manages data writing to the cache memory. This collaboration allows the storage subsystem to maintain autonomy while leveraging the server's capabilities to improve access performance, especially in large cache systems where independent processing becomes inefficient.
Data Source
AI summary
A computer system according to one preferred embodiment of the present invention has a server and a storage subsystem, wherein the server is configured to enable data write to a cache area of the storage subsystem. Further, the server manages the usages of the cache area. When storing data from the server to the cache area, the server determines whether a data-writable area exists in the cache area or not. If there is a writable area, data is stored in the writable area.