Media-Aware Content Placement Across SSD/HDD Storage Tiers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage systems fail to optimize data placement based on storage media types, leading to underutilization of high-throughput media like SSDs and inefficient data serving, especially when popularity of digital content is not considered.
Innovation Solution
A method and system that predicts the popularity of digital content and allocates it on storage media types based on their throughput, placing high-popularity content on high-throughput media like SSDs and low-popularity content on lower-throughput media like HDDs, using cluster hardware information and popularity data to optimize data storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is stored on HDDs to maximize storage capacity, then storage capacity is improved, but data throughput and access speed deteriorate
Solution Approach 1:
The patent applies local quality by assigning different storage media types (SSD vs. HDD) to different data popularity levels. High-popularity data is placed on SSDs with high throughput, while low-popularity data is placed on HDDs with lower throughput but higher capacity. This creates a heterogeneous storage system where each storage location has optimized properties for its specific data type.
Solution Approach 2:
The storage system is segmented into multiple tiers based on data popularity and media characteristics. The patent divides the storage cluster into SSD-based tiers for hot data and HDD-based tiers for cold data, allowing each segment to operate independently with optimized performance for its specific workload.
2Productivity
If data is placed on storage clusters with large available space, then storage utilization is improved, but data access speed deteriorates
Solution Approach 1:
The system performs preliminary actions by predicting data popularity before data is actually stored. Using historical access patterns and metadata analysis, the system pre-determines which storage tier each data item should occupy, preventing suboptimal placement and avoiding subsequent data migration operations.
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring actual data access patterns and comparing them with predicted popularity. This feedback loop allows the system to refine its predictions and adjust data placement decisions, ensuring optimal utilization of storage resources while maintaining fast access speeds for popular data.
3Quantity of substance
If high-capacity storage clusters are used to store all data, then storage capacity is improved, but data serving performance deteriorates
Solution Approach 1:
The patent changes the parameter of storage media type based on data characteristics. Instead of using a uniform storage medium for all data, the system dynamically selects between SSD and HDD media types based on predicted data popularity, thereby optimizing both capacity utilization and serving performance for different data sets.
Data Source
AI summary
The disclosed computer-implemented method includes accessing cluster hardware information that identifies at least two different types of storage media within a cluster and provides an indication of a respective amount of data throughput for each identified type of storage media. The method next includes accessing popularity information for digital content that is to be stored in the cluster. The popularity information indicates how often the digital content is predicted to be accessed over a specified future period of time. The method also includes allocating the digital content on the different types of storage media within the cluster according to the popularity information. Accordingly, digital content predicted to have higher popularity is placed on storage media types with higher throughput amounts, and digital content predicted to have lower popularity is placed on storage media types with lower throughput amounts. Various other methods, systems, and computer-readable media are also disclosed.