Dynamic Polling Schedule for Event Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current polling methods for data collection from servers are inefficient as they involve frequent, resource-intensive requests that waste bandwidth and CPU resources, especially when events occur infrequently, leading to delayed response times.
Innovation Solution
Implement a data processing method that uses statistical analysis to predict event occurrences and generate a targeted polling schedule based on historical event data, optimizing the frequency of poll messages by calculating probabilities and adjusting the polling schedule accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If polling is performed repeatedly at fixed intervals, then the system can detect events timely, but network bandwidth and CPU resources are wasted
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed polling interval to a dynamic polling schedule that adjusts based on predicted event probabilities. The system calculates the likelihood of events occurring at different times and automatically adjusts polling frequency accordingly, sending more frequent polls when events are more likely and reducing frequency when events are unlikely, thus optimizing resource usage while maintaining detection reliability
Solution Approach 2:
The patent changes the parameter of polling frequency from a static fixed interval to a variable schedule based on predicted event probabilities. By analyzing historical data and calculating probability distributions, the system modifies the timing parameters of polling operations to match actual event patterns, reducing unnecessary polls while ensuring timely detection when events occur
2Loss of energy
If polling frequency is reduced to save resources, then network bandwidth and CPU usage decrease, but response time for detecting events increases
Solution Approach 1:
The patent applies preliminary action by using historical event data to predict future event probabilities before actually polling for events. The system performs advance analysis of past patterns to determine the most likely times for events to occur, and schedules polling operations accordingly. This allows the system to reduce overall polling frequency while maintaining fast response times by being strategically present when events are most likely to occur
Solution Approach 2:
The patent implements feedback by continuously analyzing the outcomes of previous polling operations and event occurrences to refine the probability predictions. The system uses feedback from actual event timing patterns to improve the accuracy of future predictions, creating a self-adjusting polling schedule that optimizes both resource usage and response time over time
3Ease of manufacture
If fixed interval polling is used, then the polling schedule is simple to implement, but it cannot adapt to varying event occurrence patterns
Solution Approach 1:
The patent applies self-service by enabling the polling system to automatically generate and adjust its own schedule based on historical data analysis. Instead of requiring manual configuration to adapt to changing event patterns, the system performs self-adjustment through automated probability calculations and schedule optimization, maintaining ease of implementation while achieving high adaptability to varying event occurrence patterns
Data Source
AI summary
In an embodiment, a data processing method providing an improvement in computer efficiency in transmitting data poll messages to another computer, the method comprising: using a first computer, transmitting a first plurality of data poll messages to a second computer, receiving event data from the second computer, and storing the event data in event history storage; based upon the event data in the event history storage, calculating a first estimate of a probability of a particular event occurring in a first specified time period; in response to determining that the first estimate is greater than 0, calculating a total number of times to check for the same particular event in the same specified time period; based upon the event data in the event history storage, calculating a second estimate of a probability distribution of the same particular event during the same specified time period; creating and storing a schedule of a plurality of times at which to transmit a second plurality of data poll messages to the second computer, based upon dividing the specified time period by the second estimate; transmitting the second plurality of data poll messages to the second computer at the plurality of times specified in the schedule, receiving responsive event data from the second computer, and updating the event history storage using the responsive event data; wherein the method is performed using one or more computing devices.