Dynamic Supervision Timer Control for Client-Server Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Designing effective supervision timers for client-server systems in multi-service devices is challenging due to shared processing resources, varying latency caused by concurrent service usage, and the inability to foresee all operational scenarios, leading to overly conservative timer settings that delay detection of server failures.
Innovation Solution
A method to dynamically ascertain supervision timer values by identifying active use scenarios, measuring latency under different conditions, and adjusting timer values based on these scenarios, allowing for more accurate and context-specific timeouts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed supervision timer value is used for all services, then the system is simple to implement, but it cannot adapt to varying latency conditions caused by concurrent service usage
Solution Approach 1:
The patent implements dynamic supervision timer values that automatically adjust based on the number of active services detected in the system. The timer extends when more services are running concurrently, allowing the system to adapt to varying latency conditions without manual reconfiguration. This dynamic adjustment resolves the contradiction by making the timer versatile across different operational scenarios while maintaining automated control.
Solution Approach 2:
The patent changes the supervision timer parameter from a fixed value to a variable value that depends on the number of active services. By making the timer parameter dynamic rather than static, the system can accommodate different latency conditions caused by concurrent service usage. This parameter change enables adaptability while the system automatically determines the appropriate timer value based on current system state.
2Reliability
If the supervision timer value is set to a long time period, then the client waits longer for server responses, but server failures are detected later
Solution Approach 1:
The patent dynamically adjusts the supervision timer value based on the number of active services in the system. When fewer services are active, the timer is shorter, enabling faster detection of server failures. When more services are active, the timer is extended to accommodate increased latency. This dynamic approach resolves the contradiction by optimizing detection timing according to actual system conditions rather than using a fixed conservative value.
Solution Approach 2:
The system continuously monitors the number of active services and uses this feedback to adjust the supervision timer value. This feedback mechanism ensures that the timer is appropriately set based on current system load and latency conditions. The feedback loop allows the system to detect server failures at the optimal time, neither too early nor too late, by continuously adapting to changing operational conditions.
3Measurement precision
If the supervision timer is optimized for single-service scenarios, then it provides accurate failure detection, but it causes false timeouts when multiple services are running concurrently
Solution Approach 1:
The patent applies local quality by tailoring the supervision timer value to the specific operational context, namely the number of active services. Instead of using a universal timer value, the system adjusts the timer locally based on the current service environment. This ensures that the timer is optimized for the actual conditions present, preventing false timeouts in multi-service scenarios while maintaining precision in single-service scenarios.
Solution Approach 2:
The patent makes the supervision timer dynamic by automatically adjusting it based on the number of active services detected in the system. This dynamic behavior allows the timer to adapt from single-service optimized values to multi-service appropriate values as needed. The dynamic adjustment resolves the contradiction by enabling the system to maintain measurement precision across different service scenarios without requiring separate timer configurations for each scenario.
Data Source
AI summary
Controlling a device having a shared processing resource includes ascertaining a supervision timer value for a client service that uses the shared processing resource. The client service is caused to make a server request and a supervision timer is set to cause a timeout after a supervision timer value time period. Ascertaining the supervision timer value includes ascertaining which of a number of use scenarios represents an operation state of the device, the use scenarios being at least in part distinguished from one another by which services are presently active, including the client service. Each of the plurality of services utilizes the shared processing resource when active. One supervision timer value is selected from a number of values each associated with the client service, selection being made at least partly as a function of the ascertained use scenario. The value is retrieved from the set of stored supervision timer values.