Dynamic Window Stripe Creation for Storage Workload Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in effectively managing storage devices due to limitations in grouping and balancing usage loads, leading to inefficiencies in data reconstruction and storage space utilization, especially when the number of storage devices exceeds threshold ranges.
Innovation Solution
A method for creating a stripe in a storage system based on a dynamic window, where storage devices are selected and sequenced dynamically to avoid group limitations, ensuring balanced workloads and efficient use of storage space by discarding the concept of fixed groups and using a predetermined window to determine device sequences and correlations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If storage devices are divided into fixed groups based on threshold ranges, then management is simplified, but the number of storage devices that can be managed is limited and workload balance deteriorates
Solution Approach 1:
The patent replaces fixed group thresholds with dynamic window sizes. The window size is determined based on the total number of storage devices in the pool, allowing the system to adapt automatically as devices are added or removed. This dynamic approach eliminates the need for predetermined threshold ranges while maintaining manageable complexity through the window-based selection mechanism.
Solution Approach 2:
The patent changes the management parameter from fixed group thresholds to dynamic window sizes calculated based on total device count. By using the formula involving logarithmic scaling of the total device pool, the system automatically adjusts its management parameters to match the current scale, enabling versatility across different deployment sizes while keeping the management model consistent.
2Ease of operation
If fixed groups are used for stripe creation, then device selection is simplified, but workload balance and storage space utilization deteriorate
Solution Approach 1:
The patent implements dynamic window sizes that expand or contract based on the total number of available storage devices. This allows the system to automatically adapt the selection scope to match the current pool size, ensuring that all devices can be effectively utilized regardless of scale. The dynamic approach prevents the storage utilization problems that arise when fixed groups cannot accommodate all devices.
Solution Approach 2:
The patent moves from a two-dimensional fixed group structure to a one-dimensional dynamic window that can slide across the entire device pool. This dimensional change allows any device to be selected as the starting point for a window, enabling flexible stripe creation that can utilize any subset of devices based on current conditions, thereby improving overall storage space utilization.
3Device complexity
If fixed groups with upper and lower thresholds are implemented, then group management is easier, but devices exceeding thresholds cannot be properly grouped and workload balance deteriorates
Solution Approach 1:
The patent replaces static threshold boundaries with a dynamic window mechanism that automatically adjusts its size based on the total device pool. This eliminates the problem of devices falling outside fixed thresholds, as the window size is calculated to accommodate the current scale. The dynamic approach maintains simple management procedures while ensuring all devices are always included in the selection process, preserving workload balance.
4Quantity of substance
If the number of storage devices exceeds group thresholds, then storage capacity increases, but the ability to create stripes deteriorates
Solution Approach 1:
The patent implements a dynamic window size that scales with the total number of storage devices. As the device pool grows, the window size automatically increases according to the logarithmic scaling formula, ensuring that stripe creation remains feasible regardless of pool size. This dynamic adaptation eliminates the upper threshold limitation while maintaining efficient stripe formation through the window-based selection process.
Data Source
AI summary
Techniques involve creating a stripe based on a dynamic window. In particular, in response to a request to create a stripe in the storage system, a first storage device for creating the stripe is selected from multiple storage devices. A first extent in the first storage device is added to the stripe. A first storage device sequence associated with the first storage device is determined based on a size of a predetermined window, and the first storage device sequence includes the first storage device and a set of continuous storage devices adjacent to the first storage device in the multiple storage devices. At least one other storage device for creating the stripe is determined based on storage devices other than the first storage device in the first storage device sequence. The storage devices in the storage system can be used in a balanced manner as much as possible.