Magnetic Disk Device Dual Actuator Command Reordering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic disk devices with split actuators face challenges in efficiently reordering commands stored in queues across multiple actuators, leading to suboptimal performance and increased power consumption due to inefficient command processing and access cost calculations.
Innovation Solution
The magnetic disk device employs a system with first and second arithmetic units for each actuator, executing reordering processes and bypass arithmetic to optimize command order based on access costs, utilizing reordering and bypass arithmetic units to adjust and calculate access costs, thereby selecting low-cost commands for processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple actuators with independent queues are used, then device capacity and parallel processing capability are improved, but device complexity and difficulty of reordering management increase
Solution Approach 1:
The patent divides the command management system into separate queues for each actuator (first queue for first actuator, second queue for second actuator). Each queue independently stores commands corresponding to its specific actuator, enabling parallel processing while maintaining organized separation. This segmentation resolves the contradiction by allowing multiple actuators to operate independently without requiring complex cross-actuator coordination.
Solution Approach 2:
The patent implements a unified reordering process that can be applied to multiple actuators. The same reordering logic and access cost calculation methods are universally applied to both actuators, allowing the system to handle different actuator types and configurations through a single multi-functional framework rather than requiring separate specialized handling for each actuator.
2Loss of energy
If comprehensive reordering process is executed for all commands, then access cost optimization is improved, but processing time and computational load increase
Solution Approach 1:
The patent applies reordering processes selectively rather than universally to all commands. The system determines whether to execute reordering based on specific conditions such as command characteristics, queue state, and actuator status. This partial application of reordering optimizes access costs for commands that benefit from it while avoiding unnecessary processing time expenditure on commands where reordering would provide minimal or no benefit.
Solution Approach 2:
The reordering system incorporates self-service mechanisms where the system automatically evaluates and determines the necessity of reordering based on predefined criteria and real-time state information. The arithmetic units autonomously decide whether to perform reordering operations without requiring external intervention, balancing optimization needs against processing time constraints through self-directed decision-making.
Data Source
AI summary
According to one embodiment, a magnetic disk device includes a first disk, a second disk, a first head, a second head, a first actuator including the first head, a second actuator including the second head, a first arithmetic unit executing a first reordering process of a command stored in a first queue corresponding to the first actuator, and a second arithmetic unit executing a second reordering process of a command stored in a second queue corresponding to the second actuator, the first arithmetic unit executing the second reordering process, the second arithmetic unit executing the first reordering process.