Magnetic Disk Command Scheduling With B+ Tree Address Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing magnetic disk devices face inefficiencies in data writing and reading operations due to the time-consuming process of searching for commands in a queue to optimize access to adjacent areas on the magnetic disk, which affects operation speed.
Innovation Solution
The magnetic disk device employs a B+ tree data structure to manage unprocessed commands, allowing for efficient selection of the next command to be executed by comparing start addresses of commands in both forward and reverse directions from the current head position, ensuring faster access to adjacent areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If commands are searched in a queue to select the next command for execution, then data consistency is ensured, but operation speed decreases due to time-consuming search process
Solution Approach 1:
The patent applies preliminary action by pre-organizing commands into a B+ tree structure before execution. The B+ tree is built in advance with commands sorted by their target addresses, allowing the system to quickly navigate and select commands without searching through the entire queue sequentially. This pre-organization enables fast command selection while maintaining the FIFO queue for data consistency, thus resolving the contradiction between reliability and productivity.
2Productivity
If all commands in the queue are searched to find the optimal next command, then access efficiency is improved, but search time increases
Solution Approach 1:
The patent segments the command queue into a B+ tree structure, dividing the search space into hierarchical levels. Instead of searching all commands sequentially, the B+ tree allows the system to navigate through segments (pages) and quickly locate commands near the current magnetic disk position. This segmentation reduces search time significantly while maintaining the ability to find optimal commands for efficient access.
Solution Approach 2:
The patent transitions from a one-dimensional sequential search (queue order) to a two-dimensional search space by introducing the B+ tree structure. The tree adds an address-based dimension to command organization, enabling simultaneous consideration of both queue order (for data consistency) and address proximity (for access efficiency). This dimensional change allows fast search operations without sacrificing either criterion.
3Productivity
If commands are reordered to access closer areas first, then operation speed is improved, but command processing complexity increases
Solution Approach 1:
The patent introduces the B+ tree as an intermediary structure between the command queue and the command selection process. The B+ tree mediates between the simple FIFO queue (maintaining data consistency) and the complex reordering requirement (optimizing access speed). It provides a structured way to manage and search commands without requiring complex real-time analysis, thus reducing processing complexity while achieving speed improvements.
Data Source
AI summary
A magnetic disk includes areas to which unique addresses are respectively assigned. A memory is configured to store unprocessed commands. A controller is configured to store, in the memory, a database of a tree-shaped structure. The structure includes a plurality of nodes respectively associated with a plurality of addresses respectively specified by a plurality of unprocessed commands. Two of the plurality of nodes are associated with each other. The controller is further configured to select, of the plurality of unprocessed commands, either a command associated with an address larger than and closest to a start address or a command associated with an address smaller than and closest to the start address as a candidate command to be executed next.