SAS Topology Management via PHY Error Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In SAS networks, flaky devices or interconnects cause excessive broadcast messages and resource consumption due to frequent topology rediscovery, leading to performance issues and disruptions.
Innovation Solution
Implementing a technique within SAS devices to monitor PHY performance, detect problematic PHYs, and disable them when error thresholds are exceeded, thereby reducing the impact of flaky devices or interconnects on the network topology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a SAS expander issues BROADCAST(CHANGE) primitive to notify all SAS initiators of a domain change, then all devices are informed of the change, but complete SAS Discovery process is triggered consuming valuable resources and bandwidth
Solution Approach 1:
The patent segments the topology change notification process by introducing targeted notification to specific SAS initiators affected by the change, rather than broadcasting to all initiators. The SAS expander identifies which initiators are actually impacted by the topology change and notifies only those, dividing the notification task into specific segments rather than a blanket broadcast to all devices.
Solution Approach 2:
The patent applies partial action by performing only the necessary discovery process for affected initiators rather than requiring complete rediscovery by all initiators. When a topology change occurs, only the initiators whose paths are actually impacted need to perform discovery, reducing the overall discovery burden while still maintaining system reliability.
2Loss of information
If SAS initiators perform complete SAS Discovery process upon receiving BROADCAST(CHANGE), then topology information is updated, but valuable bandwidth and resources are consumed even for unchanged devices
Solution Approach 1:
The patent applies local quality by making the discovery process selective rather than universal. Each SAS initiator determines locally whether it is affected by the topology change before initiating discovery. Only initiators with actual topology changes perform discovery, while unaffected initiators maintain their existing topology information, creating a differentiated approach based on local conditions.
Solution Approach 2:
The patent introduces feedback mechanisms where SAS initiators receive targeted notifications about topology changes and use this feedback to determine whether discovery is necessary. The system feedbacks information about which initiators are actually affected by changes, allowing them to make informed decisions about whether to perform discovery, rather than blindly executing discovery upon receiving any broadcast.
3Adaptability or versatility
If flaky devices or interconnects remain active in SAS domain, then device connectivity is maintained, but excessive broadcast messages and resource consumption occur
Solution Approach 1:
The patent changes the operational parameters of problematic PHYs by adjusting error thresholds and connection stability criteria. Instead of maintaining all connections regardless of quality, the system monitors error rates and connection stability, and when parameters exceed acceptable thresholds, the PHY is disabled. This dynamic parameter adjustment allows the system to adapt to flaky devices while maintaining overall network productivity.
Data Source
AI summary
SAS topology is managed. Internally within a SAS device on a SAS network, a performance characteristic of a PHY of the SAS device is monitored. Internally within the SAS device, it is determined, based on the performance characteristic, that the PHY has a problem, and, based on the determination, the PHY is affected to help prevent the PHY from adversely affecting communications on the SAS network.