Lossless Network Manageability Tools for Bottleneck Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex storage area networks (SANs) face challenges in managing bottlenecks, stuck virtual channels, and lost credits, which can lead to performance issues and require effective detection and mitigation tools to ensure network reliability and efficiency.
Innovation Solution
The development of manageability tools that enable administrators to detect slow-drain and congestion bottlenecks, stuck virtual channels, and lost credits through software and hardware collaboration, with features like alerting mechanisms, automatic corrective actions, and mitigation techniques such as timeout adjustments and queue flushing, allowing for improved network management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If SAN networks are expanded to provide more interconnectivity and capacity, then network functionality and bandwidth are improved, but network complexity and difficulty of management increase
Solution Approach 1:
The patent implements a feedback mechanism where the network switch continuously monitors buffer credit levels and automatically sends flow control signals to upstream devices. This closed-loop feedback system enables the network to self-regulate traffic flow, preventing buffer overflows and credit loss without requiring manual intervention, thus resolving the contradiction between expanded network capacity and management complexity
Solution Approach 2:
The network switch performs self-diagnosis and self-correction by monitoring its own buffer credit status and automatically initiating corrective actions. The system serves itself by detecting credit loss conditions and autonomously adjusting traffic flow, eliminating the need for external management intervention and reducing the effective complexity of managing large SAN networks
2Reliability
If flow control mechanisms are implemented to prevent credit loss, then network reliability is improved, but network performance and throughput may deteriorate
Solution Approach 1:
The patent applies flow control selectively based on buffer credit levels rather than continuously. When credits are sufficient, traffic flows freely without restriction. Flow control signals are only activated when credit levels fall below thresholds, representing a partial action approach that maintains high throughput during normal operation while providing reliability protection during credit depletion scenarios
Solution Approach 2:
The system takes preliminary action by sending flow control signals before actual credit loss occurs. By proactively monitoring credit levels and initiating flow control at predetermined thresholds, the system prevents credit exhaustion and associated reliability issues while minimizing the duration and impact of flow control activation on overall network throughput
3Reliability
If buffer credits are increased to prevent overflow, then network reliability is improved, but memory usage and device resources increase
Solution Approach 1:
The patent implements dynamic buffer credit management where credit levels are adjusted based on real-time network conditions and traffic patterns. Rather than allocating fixed large buffers, the system dynamically allocates and releases credit resources, allowing the same memory pool to serve multiple purposes at different times. This dynamic approach maintains reliability by ensuring sufficient credits are available when needed while optimizing memory utilization to avoid excessive resource consumption
Data Source
AI summary
Manageability tools are provided for allowing an administrator to have better control over switches in a lossless network of switches. These tools provide the ability to detect slow drain and congestion bottlenecks, detect stuck virtual channels and loss of credits, configure hold times on edge switches to be different from hold times on core switches, and mitigate severe latency bottlenecks.