Buffer Memory Dimensioning for Real-Time Vehicle Tasks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
On-board computing systems in vehicles face challenges in precisely dimensioning buffer memories to meet the demands of real-time scheduling, which is crucial for safety-critical functions like autonomous driving and engine management, leading to inefficiencies and potential safety issues.
Innovation Solution
A method to determine the optimal size of buffer memory by analyzing writing and reading operations, determining the largest set of data elements with overlapping lifetimes, and assigning index numbers to each memory space, ensuring the buffer memory is dimensioned to meet the minimal required size for proper time scheduling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the buffer memory is dimensioned to be large enough to handle all possible data elements, then the reliability of real-time scheduling is improved, but the device complexity and resource usage increase
Solution Approach 1:
The patent applies preliminary action by analyzing the time scheduling beforehand to determine the exact buffer memory size needed. The method calculates the largest set of data elements that can be simultaneously present in the buffer based on the predetermined time scheduling, allowing the buffer to be dimensioned optimally before the system operates, thus avoiding both over-provisioning and under-provisioning.
Solution Approach 2:
The patent changes the parameter of buffer memory size from a fixed, conservative estimate to a dynamically calculated value based on the specific time scheduling parameters. By computing the buffer size as a function of the time scheduling characteristics (earliest start time, latest finish time, task periods), the system optimizes the parameter to match actual operational requirements.
2Productivity
If the buffer memory is dimensioned minimally to reduce resource usage, then the productivity is improved, but the reliability may be compromised
Solution Approach 1:
The patent performs preliminary analysis of the time scheduling to calculate the exact buffer memory requirements before system operation. By determining the largest set of data elements that can simultaneously exist in the buffer based on earliest start times and latest finish times, the system configures the buffer size precisely to meet safety requirements without wasting resources.
Solution Approach 2:
The method incorporates feedback from the time scheduling analysis to adjust the buffer memory dimensioning. The calculation uses feedback parameters including the earliest start time of any task, the latest finish time of any task, and the period of periodic tasks to continuously verify that the buffer size is sufficient for safe operation.
3Reliability
If the buffer memory size is increased to accommodate overlapping data elements, then the reliability is improved, but the use of energy increases
Solution Approach 1:
The patent optimizes the buffer memory parameter size based on the specific operational parameters of the time scheduling. By calculating the buffer size as a function of task execution times, periods, and data element lifetimes, the system determines the minimum necessary buffer capacity, avoiding energy waste from unnecessarily large buffer allocations while ensuring reliable data storage.
Solution Approach 2:
The patent applies partial action by allocating buffer memory space only for the minimum necessary duration and capacity required by the time scheduling. Rather than provisioning for maximum possible data element storage, the system allocates buffer resources partially, matching the actual operational needs derived from the time scheduling analysis.
Data Source
AI summary
A computer-implemented method dimensions a buffer memory configured to be used by multiple executable tasks. The executable tasks, according to a predetermined time scheduling, at least one of write data elements in the buffer memory or read data elements from the buffer memory. The method includes determining a plurality of writing and reading operations of the plurality of executable tasks. The method includes determining a largest set of data elements to be stored in the buffer memory according to the plurality of writing and reading operations. The method includes dimensioning the buffer memory according to the largest set of data elements.