Multiple Work Completion Messages for Hardware Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems face challenges in exposing internal diagnostics and operational statistics of hardware devices due to inadequate granularity in event logging techniques, leading to excessive irrelevant data capture, resource wastage, and increased operational costs.
Innovation Solution
Implementing multiple-completion work descriptors that allow hardware devices to generate additional completion messages with specific payloads of diagnostic information, enabling granular data capture only when needed, reducing bandwidth and storage consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If event logging techniques are used to capture hardware diagnostics and statistics, then internal hardware information becomes accessible, but excessive irrelevant data is captured leading to resource wastage
Solution Approach 1:
The patent applies local quality by making the logging behavior specific to individual work descriptors rather than applying a global logging policy. Each work descriptor can independently specify whether it should trigger completion messages, allowing selective logging of only relevant data at the granularity level of individual workloads, thus avoiding capture of excessive irrelevant data while maintaining accessibility of needed diagnostics
Solution Approach 2:
The patent segments the logging function by dividing it into per-work-descriptor decisions rather than a monolithic logging system. The completion message generation is segmented such that each work descriptor independently determines whether to generate a completion message based on its own characteristics, enabling precise control over which data is logged and reducing overall resource consumption
2Loss of information
If event logging techniques are used to capture hardware diagnostics and statistics, then internal hardware information becomes accessible, but bandwidth and storage resources are excessively consumed
Solution Approach 1:
The patent applies local quality by making the logging behavior specific to individual work descriptors rather than applying a global logging policy. Each work descriptor can independently specify whether it should trigger completion messages, allowing selective logging of only relevant data at the granularity level of individual workloads, thus avoiding capture of excessive irrelevant data while maintaining accessibility of needed diagnostics
Solution Approach 2:
The patent implements partial action by allowing only certain work descriptors to generate completion messages based on their specific characteristics, rather than having all work descriptors generate messages. This selective approach ensures that logging resources are allocated only where necessary, reducing overall data volume while maintaining diagnostic capability for relevant operations
3Loss of information
If conventional event logging is used, then hardware diagnostics can be captured, but system performance and latency are degraded
Solution Approach 1:
The patent applies local quality by making the logging behavior specific to individual work descriptors rather than applying a global logging policy. Each work descriptor can independently specify whether it should trigger completion messages, allowing selective logging of only relevant data at the granularity level of individual workloads, thus avoiding capture of excessive irrelevant data while maintaining accessibility of needed diagnostics
Solution Approach 2:
The patent implements partial action by allowing only certain work descriptors to generate completion messages based on their specific characteristics, rather than having all work descriptors generate messages. This selective approach ensures that logging resources are allocated only where necessary, reducing overall data volume while maintaining diagnostic capability for relevant operations
Data Source
AI summary
A work descriptor identifying a plurality of workflow tasks to be performed by a hardware device is generated by a host system. The work descriptor corresponds to a performance completion message generated by the hardware device in response to completing performance of the work descriptor. One or more completion indicators are added to the work descriptor. Each of the completion indicators instructs the hardware device to generate one or more additional completion messages during performance of the work descriptor in response to a trigger criterion. The work descriptor is caused to be available to the hardware device for execution.