Storage Power Control via User Access Duration Summation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed file systems face inefficiencies in power management, as prediction methods for disk access can be incorrect, leading to unnecessary reboot times and increased power consumption when disks are accessed unexpectedly.
Innovation Solution
An information processing apparatus calculates access durations for each user based on start and finish times, sums these durations for logged-in users, and determines whether to power on or off storage devices based on these calculations to optimize power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If access prediction methods are used to stop disks, then power consumption is reduced, but prediction accuracy is insufficient leading to unexpected access requests and increased reboot time
Solution Approach 1:
The system performs preliminary actions by calculating access durations and predicting future access patterns before making power management decisions. The metadata server calculates the sum of access durations for each user and uses this information to proactively determine which disks should be powered on or off, rather than reacting to access requests after they occur. This preliminary calculation enables more accurate prediction and reduces unexpected reboots.
2Loss of energy
If disks are stopped based on access prediction, then power consumption is reduced, but system reliability decreases due to incorrect predictions
Solution Approach 1:
The system implements feedback by continuously monitoring actual access patterns and using this information to refine predictions. The metadata server calculates access durations based on historical data and uses this feedback loop to improve the accuracy of power management decisions over time. This feedback mechanism ensures that disks are stopped only when predictions are highly reliable, maintaining system reliability while reducing power consumption.
3Measurement precision
If access duration calculation is performed for all users, then power management accuracy is improved, but calculation complexity increases
Solution Approach 1:
The system applies segmentation by dividing the calculation process into distinct components: the metadata server calculates access durations for each user separately, then aggregates these durations to determine overall disk usage patterns. This segmentation allows for precise per-user measurement while managing calculation complexity through modular processing. The metadata server maintains separate records for each user's access duration, enabling accurate predictions without overwhelming computational burden.
Data Source
AI summary
An information processing apparatus includes a memory and a processor coupled to the memory. The processor is configured to calculate, for each of first users, time lengths of accesses to data stored in a storage device on basis of start information on start times and finish information on finish times. The accesses are made by a node of each of the first users. The start times are times at which the respective accesses are started. The finish times are times at which the respective accesses are finished. The processor is configured to calculate, for each of second users among the first users, a first sum of the time lengths. The second users are in a logged-in state of being after login and before logout. The processor is configured to determine whether to power on the storage device on basis of the first sum.