Software-Defined Switch Routing for Distributed Storage Delays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing distributed storage systems experience increased delays and inefficiencies due to the large quantity of messages and data packets transferred between nodes, leading to higher inter-node message overheads and data forwarding amounts.
Innovation Solution
Implementing a software-defined switch with a forwarding table that maps storage partition identifiers to target storage nodes, reducing message overheads and data forwarding by directly routing requests to the appropriate nodes, and incorporating a controller to update and maintain mapping tables for real-time node status and partition information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed storage system allocates storage capacity and load to each storage server node, then storage energy and performance are improved, but a large quantity of messages and data packets are distributed and transferred on network, increasing inter-node message quantity and data forwarding amount, which increases system delay
Solution Approach 1:
The patent introduces a software-defined switch as an intermediary component between clients and storage nodes. This switch maintains a forwarding table that maps storage partition identifiers to target storage nodes, enabling direct request routing without requiring clients to communicate extensively with multiple storage nodes. The intermediary switch reduces inter-node message traffic by consolidating routing decisions, thereby decreasing system delay while preserving the distributed storage architecture's performance benefits
2Productivity
If distributed system transfers large quantity of messages and data packets on network between nodes, then information processing capacity is improved, but data forwarding amount increases, reducing information processing efficiency
Solution Approach 1:
The patent implements preliminary action by pre-establishing a forwarding table in the software-defined switch that contains mapping relationships between storage partition identifiers and target storage nodes. This allows the system to prepare routing information in advance, so that when data needs to be forwarded, the switch can immediately route packets using the pre-computed paths without performing complex lookups or decisions during actual data transfer. This reduces data forwarding overhead while maintaining high information processing capacity
Data Source
AI summary
A storage system includes a switch and a plurality of storage nodes. The switch is configured to receive a first request from a client. The first request includes an identifier of a storage partition. The switch queries an entry in a forwarding table based on the identifier to determine a target storage node in the plurality of storage nodes. The entry includes a mapping relationship between the identifier and the target storage node. The switch sends the first request to the target storage node. The target storage node is configured to receive the first request from the switch.


