Process Activeness Check via Progress Logs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques in multi-node computer systems fail to detect and address issues in process activeness effectively, leading to late detection of problems and adverse performance degradation, as they rely on inferring node failures rather than monitoring process activity dynamically.
Innovation Solution
Implementing a mechanism where processes update progress logs, allowing counterpart processes to monitor activity, determine issues such as CPU starvation, data corruption, or deadlocks, and trigger restoration strategies, including rebooting or resetting nodes, to address problems promptly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If heartbeat mechanism is used to detect node failures, then node failure detection is enabled, but process-level activeness monitoring is not provided and detection is too late
Solution Approach 1:
The patent segments the node monitoring function into two levels: node-level heartbeat monitoring (existing) and process-level activeness monitoring (new). Each process maintains its own activeness log, allowing individual process monitoring independent of the node heartbeat mechanism. This segmentation enables detection of process-level issues before they cause node failure.
Solution Approach 2:
The patent implements preliminary monitoring of process activeness by continuously logging process activities (database operations, message processing, etc.) before failures occur. The activeness log records process state changes in advance, enabling early detection of degradation patterns and proactive intervention before the process becomes unresponsive or causes node failure.
2Device complexity
If processes are monitored using heartbeat messages only, then simple node availability is tracked, but detailed process activeness and performance issues remain undetected
Solution Approach 1:
The patent introduces an activeness log as an intermediary data structure between the process and monitoring mechanisms. The process updates its activeness log to record detailed operations (database reads/writes, message processing, computational tasks), and monitoring processes query this log to assess activeness. This intermediary enables precise measurement of process activeness without requiring complex direct instrumentation of the monitored process.
3Ease of operation
If no dynamic process monitoring is implemented, then system operation is simple, but performance degradation goes unnoticed until failure occurs
Solution Approach 1:
The patent implements self-service monitoring where each process automatically maintains its own activeness log, recording its operational status without external intervention. The process itself generates the monitoring data by logging its activities, eliminating the need for complex external monitoring infrastructure. Other processes or monitoring mechanisms can then query these self-maintained logs to assess activeness and detect performance degradation early.
Data Source
AI summary
Described herein are techniques for dynamically monitoring process activeness of processes running on a computing node. Problems affecting processes to serve their designated functions on the computing node can be relatively quickly detected and dealt with, thereby making restoring process activeness on the computing node much more quickly than otherwise.