Storage Controller Logic Paths for I/O Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Storage area network controllers face high latency and processing burdens due to the generic nature of their CPUs, which are not specialized for specific input-output operations, leading to inefficiencies in handling high volumes of requests.
Innovation Solution
A storage controller with integrated circuits featuring multiple logic paths, each comprising specific logic blocks, allows for the identification and execution of input-output commands directly in hardware, reducing reliance on general-purpose processors and minimizing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a general-purpose CPU is used to handle all storage controller functionality, then the system maintains flexibility and ease of programming, but processing latency increases and productivity decreases under high request volumes
Solution Approach 1:
The storage controller is segmented into multiple independent logic paths, each dedicated to handling specific input-output operations. This segmentation allows parallel processing of different command types (read, write, delete, etc.) simultaneously, thereby increasing processing throughput while maintaining the flexibility of a software-based control plane for configuration and management.
Solution Approach 2:
An intermediary command identification mechanism is introduced that sits between the incoming packet and the execution logic. This intermediary identifies the command type and routes it to the appropriate logic path, enabling specialized hardware acceleration for common operations while preserving the ability to handle diverse commands through a unified interface.
2Adaptability or versatility
If a general-purpose CPU is used for storage controller operations, then the system maintains adaptability to different operations, but processing speed decreases due to lack of specialization
Solution Approach 1:
The storage controller implements a universal architecture with multiple logic paths that can collectively handle various input-output operations. Each logic path is specialized for specific operations, but the system as a whole maintains versatility by supporting multiple operation types through the array of specialized paths, achieving both speed and adaptability.
Solution Approach 2:
The system dynamically selects which logic path to use based on the command type identified in the incoming packet. This dynamic routing ensures that each operation is processed by the most appropriate specialized logic path, maximizing processing speed while maintaining adaptability to different operation types through flexible command routing.
3Productivity
If more logic paths are added to handle high request volumes, then processing capacity increases, but device complexity increases
Solution Approach 1:
The integrated circuit is segmented into modular logic paths, where each path is a self-contained unit handling a specific operation type. This modular segmentation allows the system to scale processing capacity by adding or activating specific logic paths as needed, rather than requiring a monolithic complex circuit, thereby managing device complexity while increasing request handling capacity.
Data Source
AI summary
A storage controller for a storage area network that implements input-output commands in hardware is provided. Specific input-output commands are defined, including: a READ command, a WRITE command, a COPY command and a WRITE MIRROR command, all of which are implemented in hardware. A virtual logical unit table can be provided that enables the storage controller to implement a virtual storage management overlay function on top of a raw physical storage function.