Demand-Based SSD Architecture for Transparent Remote Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage devices with fixed capacities fail to adapt to fluctuating storage needs, leading to inefficiencies and disruptions when capacity requirements change, and disaggregated storage solutions introduce network processing overhead and complexity.
Innovation Solution
A demand-based storage device (SoD-SSD) that provides scalable storage capacity by leveraging remote storage devices over a network, allowing transparent access beyond physical limits and offering features like QoS, RAID, and data security, while appearing as a traditional DAS SSD to the host.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a traditional storage device with fixed storage capacity is used, then the device structure is simple and reliable, but it cannot adapt to changing capacity requirements of the host
Solution Approach 1:
The storage system is segmented into a local storage device (first storage device) and remote storage devices (second storage devices). The local device handles control logic and command translation, while remote devices provide actual storage capacity. This segmentation enables the system to scale storage capacity independently without increasing the complexity of the host interface.
Solution Approach 2:
The first storage device acts as an intermediary between the host and second storage devices. It translates storage commands from the host into appropriate commands for the remote storage devices, masking the complexity of the distributed storage system from the host while enabling flexible capacity expansion.
2Adaptability or versatility
If disaggregated storage solutions are implemented, then storage capacity can be scaled flexibly, but network processing overhead and system complexity increase
Solution Approach 1:
The first storage device serves as a mediator that handles network communication and command translation. It receives commands from the host over a data communication bus, translates them into appropriate commands for remote storage devices accessed over network fabric, and manages the complexity of network processing independently from the host.
Solution Approach 2:
The first storage device creates a virtual copy of the remote storage capacity, presenting it to the host as if it were locally attached. This virtualization allows the host to access remote storage through standard interfaces without being aware of the network infrastructure, thereby reducing the perceived complexity.
3Adaptability or versatility
If remote storage devices are used to provide on-demand capacity, then storage flexibility improves, but access time and response delay may increase
Solution Approach 1:
The system performs preliminary actions by pre-establishing command translation rules and maintaining ready-state connections to remote storage devices. When the host issues storage commands, the first storage device can immediately translate and forward them without delay, reducing the time penalty of remote access.
Solution Approach 2:
The first storage device as an intermediary optimizes the access path by caching frequently accessed data locally or maintaining buffer zones. It manages data transfer between the host and remote storage devices efficiently, minimizing the time impact of remote access while maintaining the flexibility benefits.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for demand-based storage are disclosed. A first storage device is configured to be coupled to a first computing device over a first link. The first storage device includes a storage medium and a processing circuit connected to the storage medium. The processing circuit may be configured to: receive a first request for a first storage capacity; transmit a second request for allocating at least a portion of the first storage capacity on a second storage device configured to communicate with the first storage device over a second link; receive a first storage command from the first computing device; generate a second storage command based on the first storage command; and transmit the second storage command to the second storage device for execution of the second storage command by the second storage device.