Micro File Dispatch Queues for I/O Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing file systems face inefficiencies due to the large number of micro files, which cause internal fragmentation, waste of magnetic disk space, and increased I/O operations, leading to reduced overall I/O efficiency.
Innovation Solution
A method and apparatus that determine and group micro file invoking request messages into dispatch queues based on continuous physical sector addresses, allowing for combination and deletion processing to reduce the number of I/O operations and optimize storage space utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate data blocks are allocated to micro files, then each file can be accessed independently, but internal fragments are caused and magnetic disk space is wasted
Solution Approach 1:
The patent combines multiple micro files into a single directory entry, merging their access requests. When micro files are stored in the same data block, their invoking requests are merged into one I/O operation, eliminating separate access overhead and reducing disk space waste from internal fragmentation while maintaining independent file access through the combined entry.
2Ease of operation
If separate data blocks are allocated to micro files, then file access is simplified, but a large quantity of I/O operations are caused and system invoking is excessive
Solution Approach 1:
The patent merges invoking requests for multiple micro files into a single combined I/O operation. By storing micro files in the same data block and combining their access requests, the system reduces the number of I/O operations and system invocations while maintaining simplified file access through the combined directory entry.
3Speed
If frequent read/write of micro files is performed, then data accessibility is improved, but context switching increases and CPU and memory resources are occupied
Solution Approach 1:
The patent combines multiple micro file access operations into a single I/O operation through the merged directory entry. This merging approach maintains fast data accessibility by allowing quick access to all micro files in the combined entry while reducing context switching and CPU/memory resource consumption by eliminating repeated system invocations.
4Device complexity
If a fixed data block size is used during formatting, then storage structure is simplified, but internal fragments are caused and I/O efficiency is reduced
Solution Approach 1:
The patent maintains the simplicity of fixed data block sizing while improving I/O efficiency by merging multiple micro file access requests into a single operation. The fixed block structure remains intact, but the combined directory entry allows multiple files within the same block to be accessed through one I/O operation, eliminating the efficiency loss from internal fragmentation.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Embodiments of the present invention relate to a method and an apparatus for accessing a file, and a storage system. The method includes: determining L first-file invoking request messages; separately adding, according to physical sector information in the L first-file invoking request messages, the L first-file invoking request messages to M first-file dispatch queues; and processing a first-file invoking request message included in each first-file dispatch queue, to obtain an invoking request message of each first-file dispatch queue, where the invoking request message is used to access a first file. By means of the method and apparatus for accessing a file and the storage system in the embodiments of the present invention, a corresponding dispatch queue is allocated to an invoking request message of a micro file, so that combination processing can be performed on invoking request messages of multiple micro files, and a quantity of times of invoking a micro file is reduced. In this way, it can be avoided that a quantity of I/O times is increased and excessive memory resources are occupied because micro file invoking requests increase, and a throughput of I/O requests is increased, thereby increasing overall I/O efficiency of a file system.