Electronic Storage Mapping for Host-Free Workload Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory devices face challenges in efficiently managing data storage and retrieval across multiple storage devices without overburdening the host device, leading to potential bottlenecks and reduced device lifetime.
Innovation Solution
An electronic device comprising a host device and multiple storage devices that share status information and redirect memory operations based on shared status information, with each storage device storing mapping information of logical and physical block addresses to distribute workload and improve load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the host device directly manages data storage and retrieval across multiple storage devices, then the host device can control data operations, but the host device becomes overburdened and creates bottlenecks
Solution Approach 1:
The patent introduces mapping information as an intermediary layer between the host device and multiple storage devices. The mapping information contains logical block addresses (LBAs) and physical block addresses (PBAs) that enable indirect addressing and data management. This intermediary structure allows the host device to issue commands without directly managing the complexity of multiple storage devices, thereby reducing host burden while maintaining control capability.
Solution Approach 2:
The patent segments the address space into logical block addresses (LBAs) and physical block addresses (PBAs). The mapping information divides the storage management function into separate components: LBA management for the host interface and PBA management for the actual storage operations. This segmentation allows independent optimization of each layer and prevents the host device from being overwhelmed by the complexity of direct multi-device management.
2Quantity of substance
If multiple storage devices are used to increase storage capacity, then data storage capability is improved, but workload distribution and load balancing become complex
Solution Approach 1:
The mapping information structure serves multiple functions simultaneously: it acts as an address translation layer, a workload distribution mechanism, and a load balancing coordinator. The same mapping information that enables data retrieval also facilitates intelligent data placement and redistribution across storage devices. This multi-functionality reduces the need for separate complexity-managing components while maintaining storage capacity expansion.
Solution Approach 2:
The patent implements feedback mechanisms where mapping information is dynamically updated based on storage device status and workload conditions. The system monitors storage device performance and uses this feedback to intelligently redistribute data and update mapping relationships, achieving automatic load balancing without complex external management. This feedback-driven approach simplifies workload distribution while maintaining optimal performance across multiple storage devices.
3Device complexity
If data is stored in a single storage device, then management is simple, but device lifetime is reduced due to concentrated wear
Solution Approach 1:
The patent performs preliminary data placement and mapping configuration before actual storage operations occur. The mapping information is pre-established to distribute data across multiple storage devices, preventing concentrated wear from the outset. This preliminary distribution strategy ensures that write operations are automatically spread across multiple devices, extending overall system lifetime while maintaining management simplicity through the unified mapping interface.
Solution Approach 2:
The patent dynamically changes the mapping parameters (LBA to PBA relationships) to redistribute data across storage devices based on their current status and wear levels. By adjusting the mapping information, the system can redirect future write operations to less-worn devices, thereby equalizing wear patterns and extending the operational lifetime of the storage system while keeping the management interface simple and unchanged.
4Reliability
If the host device intervenes in data redistribution across storage devices, then optimal data placement can be achieved, but system overhead increases
Solution Approach 1:
The patent enables the storage system to perform self-service data redistribution through automatic mapping updates. The mapping information structure allows storage devices to autonomously reposition data based on predefined policies and current system state, without requiring host device intervention. This self-service mechanism achieves optimal data placement while eliminating the time overhead associated with host-mediated redistribution operations.
Data Source
AI summary
Disclosed is a method of operating an electronic device which includes a first storage device, a second storage device, and a host device. The method includes providing a first request indicating a memory operation of target data to the first storage device, providing a second request indicating a redirection of the memory operation to the second storage device based on the first request, performing the memory operation based on the second request, providing the first storage device with a first completion response indicating that the memory operation is performed, and providing a second completion response to the host device based on the first completion response. The first storage device stores first mapping information of a logical block address (LBA) of the target data and a physical block address (PBA) indicating a physical block of the second storage device, in which the target data are stored.