Attachable Compute Modules for Local Storage Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face challenges in efficiently processing large amounts of data due to centralized data processing delays and resource consumption, and manufacturing multiple devices for different applications increases costs and complexity.
Innovation Solution
A modular compute resource architecture for storage devices that allows attachable compute modules to be connected, enabling storage devices to process data directly, with a storage device controller managing control and communication through command extensions and interfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple storage devices are manufactured for different applications, then application-specific performance is improved, but manufacturing costs and complexity increase
Solution Approach 1:
The storage device is designed with a universal architecture that can perform multiple applications through attachable compute modules. The storage device controller and interface can work with different types of compute modules (AI/ML, database, video processing, etc.), allowing a single base device to serve multiple purposes rather than manufacturing separate devices for each application
Solution Approach 2:
The compute resources are segmented into separate attachable modules that can be independently designed, manufactured, and attached to the storage device. This allows the compute functionality to be divided into specialized units (AI/ML module, database module, video processing module, etc.) that can be selectively attached based on application needs, reducing the complexity of manufacturing multiple complete devices
2Device complexity
If centralized data processing is used, then system architecture is simplified, but processing delays and resource consumption increase
Solution Approach 1:
The processing function is segmented between the storage device controller and attachable compute modules. The controller handles storage management and data retrieval, while specialized compute modules handle application-specific processing tasks, enabling parallel operation and reducing bottlenecks associated with centralized processing
Solution Approach 2:
The attachable compute modules serve as intermediaries between the storage medium and the host system. They receive data from the storage medium, perform processing operations locally, and return results to the host, eliminating the need for all data to be transferred to and processed by a centralized system
Data Source
AI summary
A storage device may include a storage medium, a storage device controller coupled to the storage medium, a host interface coupled to the storage device controller, and an attachable module interface configured to connect an attachable compute module to the storage device controller. The attachable module interface may include a data interface, a side-band interface, and/or a power interface. The attachable module interface may include a connector configured to connect the attachable compute module to the storage device controller. The storage device may include an enclosure having an opening configured to enable the attachable compute module to be connected to the attachable module interface through the opening. The storage device controller may be configured to utilize one or more resources of the attachable compute module. The storage device controller may be configured to communicate with the attachable compute module through one or more command extensions of a storage protocol.