Service Traffic Allocation Using Indicator Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for allocating traffic to multiple services based on historical data lead to excessive concentration on services nearing saturation, resulting in low traffic utilization for services that have not reached saturation.
Innovation Solution
A method involving indicator sampling values and recommendation probabilities to determine traffic allocation proportions, using a combination of recommendation constraints, constraint compliance times, and weighted summations to optimize traffic distribution among services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traffic is allocated based on service historical data (historical exposure rate and historical conversion rate), then services with higher historical conversion rates receive more traffic, but traffic becomes excessively concentrated on services that already reach or are about to reach traffic saturation, resulting in excessively low traffic utilization
Solution Approach 1:
The patent changes the parameter basis from historical data to indicator sampling values. By using sampling values obtained through random sampling of indicator data (such as conversion rate, exposure rate, etc.), the system dynamically adjusts traffic allocation proportions based on current service performance rather than static historical averages, thereby preventing traffic saturation and improving overall traffic utilization
Solution Approach 2:
The patent introduces a feedback mechanism where indicator sampling values are continuously collected and used to adjust traffic allocation. The system calculates traffic allocation proportions based on the sampled indicator values and uses this feedback to dynamically redistribute traffic, ensuring that services with better sampled performance receive more traffic while preventing concentration on saturated services
Data Source
AI summary
A service data processing method, apparatus, and computer-readable storage medium for allocating traffic among multiple services. The method includes acquiring N services and M indicator sampling values corresponding to these services, where M and N are integers greater than 1. Based on indicator sampling values satisfying a recommendation constraint, recommendation probabilities are determined for each service. A first traffic allocation proportion is calculated based on the ratio of each service's recommendation probability to the total accumulated recommendation probability. A second traffic allocation proportion is then determined using a weighted summation of the first allocation proportion and a reference allocation proportion. The N services are pushed on a platform according to this second traffic allocation proportion, with the reference allocation proportion being based on indicator sampling values not satisfying the recommendation constraint.