Parallel Torus Network Interconnect Bandwidth Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional torus network topologies face limitations such as limited bandwidth and increased traffic latency due to link congestion or failure, which affects quality of service and class of service, and redundant links reduce bandwidth while attempting to mitigate these issues.
Innovation Solution
A parallel tori interconnect system where multiple tori networks are connected, allowing a host to distribute data traffic across multiple nodes in different tori, each with its own bandwidth and quality of service, thereby increasing overall network bandwidth and reducing the impact of link failures or congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If hosts connect to more nodes in the torus to increase bandwidth, then bandwidth availability improves, but bandwidth is stolen from other nodes and hosts causing network resource contention
Solution Approach 1:
The invention divides the network into multiple independent torus fabrics (e.g., torus 0, torus 1, torus 2) where each fabric operates separately. Hosts can connect to nodes across different torus fabrics, effectively segmenting the bandwidth resources. This allows a host to access bandwidth from multiple torus fabrics simultaneously without stealing bandwidth from other hosts in the same fabric, thus resolving the resource contention issue while maintaining high bandwidth availability.
2Quantity of substance
If links are added to increase bandwidth capacity, then overall bandwidth improves, but device complexity and cost increase
Solution Approach 1:
The invention merges multiple torus fabrics into a single logical network system where hosts can utilize resources from any fabric. Instead of adding more links within a single torus fabric, the system combines multiple existing torus fabrics and allows flexible routing across them. This approach increases bandwidth capacity by utilizing parallel paths across fabrics without requiring additional physical links within each fabric, thus avoiding increased device complexity.
3Reliability
If traffic is rerouted around failed or congested links to maintain connectivity, then network reliability improves, but traffic latency increases
Solution Approach 1:
The invention introduces multiple torus fabrics as intermediary networks between source and destination nodes. When a link fails or becomes congested in one fabric, traffic can be routed through alternative paths in other fabrics, using them as intermediary routes. This maintains network connectivity while providing shorter alternative paths compared to traditional rerouting within the same fabric, thus reducing latency while maintaining reliability.
Data Source
AI summary
A system and method for optimizing a flow of data traffic are provided. A plurality of tori are connected in a parallel tori interconnect. Each torus includes a plurality of nodes. The nodes in the torus are interconnected using links. A host in the network is connected to a subset of nodes where nodes in the subset are associated with different tori. The host transmits the packets to the parallel tori interconnect by selecting a node the subset of nodes. The packets are transmitted using links between from the node to the plurality of nodes in the torus, but not between the plurality of tori.