Dynamic Synthetic Transaction Scheduling for E-Business Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for scheduling synthetic transactions in e-business systems face challenges in determining the optimal frequency, leading to either excessive traffic and performance degradation or delayed detection of network element faults due to insufficient frequency.
Innovation Solution
Implementing a dynamic scheduling mode that overrides pre-specified schedules for synthetic transactions upon detecting issues, allowing monitoring agents to run additional transactions immediately and correlate results to determine if problems are localized within the network topology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If synthetic transactions are generated at high frequency, then fault detection speed is improved, but network traffic increases and system performance degrades
Solution Approach 1:
The patent implements dynamic scheduling of synthetic transactions that adjusts transaction frequency based on system conditions. During normal operation, transactions run at low frequency to minimize impact. When anomalies are detected, the system automatically increases transaction frequency to rapidly diagnose issues, thus adapting the monitoring intensity to actual system needs rather than using a fixed schedule
Solution Approach 2:
The system continuously monitors performance metrics from synthetic transactions and uses this feedback to adjust future transaction scheduling. When performance degradation or faults are detected, the system responds by increasing transaction frequency to gather more diagnostic data. This closed-loop feedback mechanism ensures that high-frequency transactions are only executed when actually needed for fault detection
2Productivity
If synthetic transactions are generated at low frequency, then system performance is maintained, but fault detection is delayed
Solution Approach 1:
The monitoring system transitions from static low-frequency scheduling to dynamic scheduling that can scale intensity based on conditions. The base frequency remains low to preserve system performance, but the system maintains the capability to rapidly increase frequency when triggered by performance anomalies or error conditions, ensuring both efficiency and responsiveness
3Reliability
If synthetic transactions are run frequently across all network zones, then comprehensive monitoring is achieved, but unnecessary traffic is generated
Solution Approach 1:
The patent divides the network into multiple zones with monitoring agents deployed in each zone. Instead of uniformly executing transactions across all zones, the system selectively activates transactions in specific zones based on where anomalies are detected or where they are most likely to occur. This localized approach maintains comprehensive monitoring capability while minimizing unnecessary traffic in healthy zones
Solution Approach 2:
The monitoring system is segmented into multiple independent zones with dedicated agents, allowing selective execution of transactions in specific segments rather than forcing all agents to execute transactions uniformly. This segmentation enables the system to concentrate monitoring resources where needed while leaving other segments in a low-activity state
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method and apparatus is provided for monitoring operations of a specified transaction server that has an associated network topology. One embodiment comprises the steps of defining a plurality of zones within the network topology, and assigning one or more monitoring agents to each of the zones, wherein each agent is adapted to selectively run synthetic transactions with the specified server. The method further comprises monitoring results of successive synthetic transactions carried out by the agents, in order to detect any errors associated with the successive transactions. In response to detecting a performance or an availability problem, selectively, that is associated with a particular synthetic transaction run by a particular one of the agents, one or more agents is dynamically scheduled to run synthetic transactions, wherein each scheduled transaction has a specified relationship with the particular transaction.