Cloud Disk Scheduling Using Compression Rate Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud disk scheduling in cloud computing architectures fails to consider differences in data compression rates among users, leading to unbalanced resource allocation, excessive disk wear, and increased risk of data damage due to unbalanced compression rates across storage clusters.
Innovation Solution
A cloud disk scheduling method that predicts the compression rate of each storage cluster based on historical user data and current storage status to balance compression rates by selecting an appropriate target storage cluster for new cloud disks, using a negative feedback mechanism to adjust predictions dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data compression technology is introduced in the writing path to reduce storage space requirements, then storage efficiency is improved, but compression rates become unbalanced across storage clusters leading to excessive disk wear and increased risk of data damage
Solution Approach 1:
The system dynamically adjusts cloud disk scheduling decisions based on real-time compression rate monitoring. When a storage cluster's compression rate exceeds a threshold, the system dynamically redirects new cloud disks to other clusters with lower compression rates, preventing excessive disk wear and maintaining data safety while preserving storage efficiency benefits
Solution Approach 2:
The system implements a feedback mechanism that continuously monitors compression rates of storage clusters and uses this information to adjust scheduling decisions. The feedback loop ensures that compression rate imbalance is detected and corrected, preventing the accumulation of excessive disk wear and maintaining system reliability
2Device complexity
If cloud disks are allocated to storage clusters without considering compression rate differences, then resource allocation is simplified, but unbalanced compression rates cause excessive disk wear and reduced system stability
Solution Approach 1:
The system performs preliminary assessment of compression rates and storage cluster status before allocating new cloud disks. By predicting future compression rates and identifying clusters that would experience excessive compression, the system proactively directs new disks to appropriate clusters, preventing premature disk wear and extending disk service life
Solution Approach 2:
The system changes scheduling parameters based on compression rate thresholds and storage cluster status. By adjusting allocation decisions according to real-time compression rate data, the system optimizes disk distribution to prevent excessive wear on any single cluster, thereby extending the operational duration of storage devices
3Speed
If storage resources are allocated without balancing compression rates, then allocation speed is increased, but unbalanced compression rates increase the risk of storage cluster failures
Solution Approach 1:
The system implements real-time feedback monitoring of compression rates across storage clusters and adjusts allocation decisions accordingly. This feedback mechanism ensures that new cloud disks are distributed to clusters with appropriate compression rates, maintaining cluster stability while preserving fast allocation speeds through automated decision-making
Solution Approach 2:
The allocation system dynamically adapts to changing compression rate conditions across storage clusters. By continuously monitoring and responding to compression rate variations, the system maintains balanced resource distribution that prevents cluster failures while preserving allocation speed through automated, real-time decision-making
Data Source
AI summary
Embodiments of the present disclosure provide a cloud disk scheduling method, device and a storage medium based on an elastic block storage service. The method includes: obtaining historical cloud disk usage status information of a user to whom a cloud disk to-be-created belongs and current storage status information of each storage cluster, and predicting a predicted compression rate of each storage cluster in a case that the cloud disk to-be-created is created in different storage clusters; and determining a target storage cluster from each storage cluster according to the predicted compression rate of each storage cluster, such that a difference among compression rates of all storage clusters is minimized after the cloud disk to-be-created is created in the target storage cluster.