NIC Performance Notification Packet Assembly for Network Overhead Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network performance monitoring methods are inefficient as they often require frequent status queries and logging, leading to high communication overhead and delayed alerts, especially when critical conditions are reached.
Innovation Solution
The implementation of a Network Interface Controller (NIC) that generates performance indicators, classifies them by severity, and assembles notifications into packets only when thresholds are exceeded, which are then sent over the network, utilizing a Scheduler to control packet transmission based on size, time elapsed, and urgency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequent status queries and logging are performed for network performance monitoring, then monitoring coverage and detection capability are improved, but network communication overhead increases and system efficiency deteriorates
Solution Approach 1:
The patent extracts and transmits only the critical performance indication data that exceeds predefined thresholds, rather than logging all status queries. The NIC identifies and sends only significant performance events (such as errors, congestion, or threshold violations) to the monitoring server, filtering out normal operational data that does not require attention.
Solution Approach 2:
The system changes the parameter of data transmission by using event-driven triggering based on severity thresholds. Instead of continuous periodic logging, performance indications are transmitted only when specific parameter conditions are met (e.g., error rates exceed threshold, latency spikes), dynamically adjusting transmission frequency based on actual network conditions.
2Loss of information
If all performance indications are transmitted to the monitoring server, then complete monitoring data is obtained, but network load and server processing burden increase
Solution Approach 1:
The patent applies local quality by implementing threshold-based filtering at the source (NIC). Different performance indications are evaluated against severity thresholds locally, and only those meeting the criteria are transmitted. This creates a quality filter that ensures transmitted data is both complete (all significant events) and concise (no redundant normal operations).
3Speed
If performance notifications are sent immediately when thresholds are exceeded, then real-time alerting is achieved, but network traffic spikes occur
Solution Approach 1:
The system uses periodic batching combined with event triggering. Performance indications are collected and evaluated periodically, and notifications are sent in batches when threshold violations occur, rather than immediate individual transmissions. This periodic action smooths network traffic while maintaining timely alerting for critical events.
4Loss of time
If detailed performance logging is performed at high frequency, then high timing definition monitoring is achieved, but system resource consumption increases
Solution Approach 1:
The patent applies partial action by monitoring all performance parameters at high frequency but transmitting only the partial set of data that exceeds thresholds. The NIC continues to collect and time-stamp all performance indications with high timing precision, but selectively transmits only significant events, achieving high timing definition where needed while improving overall system efficiency.
Data Source
AI summary
A method includes, in a Network Interface Controller (NIC) that communicates over a network, generating indications pertaining to a performance of the NIC. The indications are classified with respect to severity. At least some of the indications, for which the severity exceeds a predefined severity threshold, are assembled in performance notification packets. The performance notification packets are sent over the network.