Real-Time Task Scheduling with Adaptive Security Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Real-time computing systems face challenges in meeting temporal constraints while ensuring security, as security mechanisms introduce time overhead and are not adequately addressed in existing technologies, particularly in critical applications like nuclear power and aircraft control.
Innovation Solution
A method for scheduling task instances in real-time systems that assigns parameters including scheduled times and security services based on time differences, using MAC functions for integrity and authentication, and employing the Earliest Deadline First rule to prioritize tasks, thereby balancing security and time constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If security mechanisms are implemented in real-time systems, then security level is improved, but time overhead increases and temporal constraints may not be met
Solution Approach 1:
The patent applies dynamics by making the security service level adaptive rather than static. The scheduling method dynamically adjusts the security service applied to each task instance based on the calculated time difference between scheduled completion and current time. When time is充裕, higher security services are applied; when time is tight, security services are reduced to meet temporal constraints. This dynamic adjustment resolves the contradiction between maintaining high security and meeting time deadlines.
Solution Approach 2:
The patent changes the parameter of security service level from a fixed value to a variable that depends on time constraints. By calculating the time difference and using it to determine the appropriate security service, the system transforms security from a static requirement into a flexible parameter that can be adjusted based on real-time conditions, thereby balancing security needs with temporal constraints.
2Productivity
If computing resources are increased to meet temporal constraints, then productivity is improved, but system complexity and cost increase
Solution Approach 1:
The patent applies partial action by providing security services selectively rather than uniformly to all task instances. Based on the time difference calculation, the system applies appropriate security services only when time permits, rather than applying maximum security to every task. This partial application of security services reduces the overall computational burden and resource requirements while still maintaining adequate security where time allows.
3Reliability
If security services are applied to all task instances, then security level is improved, but time overhead increases and may cause deadline violations
Solution Approach 1:
The patent applies local quality by making security service application non-uniform across different task instances. Instead of applying the same security level to all tasks, the system calculates the time difference for each task instance and applies security services locally based on individual time availability. This allows some tasks to receive enhanced security while others receive minimal or no security, optimizing the balance between security coverage and time consumption.
Data Source
Figure 1~2
Figure 3a~3c
AI summary
This invention relates to a method for scheduling a sequence of task instances in a real-time computing system, comprising assigning parameters to each instance, said parameters comprising a scheduled time; and, for each current instance to be scheduled (S1); - determining (S2) a time difference between the scheduled time assigned to said current instance and a current time; - assigning (S3) a security service to at least one instance of said sequence from said current instance, based on said time difference.