• Journal of KIISE:Computing Practices and Letters
• /
• v.5 no.4
• /
• pp.467-478
• /
• 1999

실시간 태스크의 최악 실행시간을 예측할 때 과예측이 발생하는 원인은, 첫째 프로그램의 동적인 최악 실행 행태를 정적으로 분석하는 것이 근본적으로 어렵기 때문이며, 둘째 최근의 RISC 형태 프로세서에 포함되어 있는 파이프라인 실행 구조와 캐쉬 등이 그러한 정적 분석을 더욱 어렵게 만들기 때문이다. 그런데 기존의 연구에서는 각각의 과예측 원인을 해결하기 위한 방법에 대해서만 언급하고 있을 뿐 분석의 정확도에서 각 원인이 차지하는 비중에 대해서는 언급하고 있지 않다. 이에 본 연구에서는 최악 실행시간 예측시 과예측을 유발하는 원인들, 즉 분석 요소들의 영향을 정량적으로 조사함으로써 기존의 최악 실행시간 분석 기법들이 보완해야 할 방향을 제시하고자 한다. 본 연구에서는 실험이 특정 분석 기법에 의존하지 않도록 하기 위하여 시뮬레이션 방법에 기반한다. 이를 위해 분석 요소별 스위치가 포함된 MIPS R3000 프로세서를 위한 시뮬레이터를 구현하였는데, 각 스위치는 해당 분석 요소에 대한 분석의 정확도 수준을 결정한다. 모든 스위치 조합에 대해서 시뮬레이션을 반복 수행한 다음 분산 분석을 수행하여 어떤 분석 요소가 가장 큰 영향을 끼치는지 고찰한다.Abstract Existing analysis techniques for estimating the worst case execution time (WCET) of real-time tasks still suffer from significant overestimation due to two types of overestimation sources. First, it is unavoidably difficult to predict dynamic behavior of programs statically. Second, pipelined execution and caching found in recent RISC-style processors even more complicate such a prediction. Although these overestimation sources have been attacked in many existing analysis techniques, we cannot find in the literature any description about questions like which one is most important. Thus, in this paper, we quantitatively analyze the impacts of overestimation sources on the accuracy of the worst case timing analysis. Using the results, we can identify dominant overestimation sources that should be analyzed more accurately to get tighter WCET estimations. To make our method independent of any existing analysis techniques, we use simulation based methodology. We have implemented a MIPS R3000 simulator equipped with several switches, each of which determines the accuracy level of the timing analysis for the corresponding overestimation source. After repeating simulation for all of the switch combinations, we perform the variance analysis and study which factor has the largest impact on the accuracy of the predicted WCETs.

• Journal of Computing Science and Engineering
• /
• v.8 no.4
• /
• pp.215-227
• /
• 2014

In modern computer architectures, caches are widely used to shorten the gap between processor speed and memory access time. However, caches are time-unpredictable, and thus can significantly increase the complexity of worst-case execution time (WCET) analysis, which is crucial for real-time systems. This paper proposes a time-predictable two-level scratchpad-based architecture and an ILP-based static memory objects assignment algorithm to support real-time computing. Moreover, to exploit the load/store latencies that are known statically in this architecture, we study a Scratch-pad Sensitive Scheduling method to further improve the performance. Our experimental results indicate that the performance and energy consumption of the two-level scratchpad-based architecture are superior to the similar cache based architecture for most of the benchmarks we studied.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.134-142
• /
• 2021

This paper proposes a probabilistic method in analyzing timing measurements to determine the periodicity of real-time tasks. The proposed method fills a gap in existing techniques, which either concentrate on the estimation of worst-case execution times, or do not consider the stochastic behavior of the real-time scheduler. Our method is based on the Z-test statistical analysis which calculates the probability of the measured period to fall within a user-defined standard deviation limit. The distribution of the measured period should satisfy two conditions: its center (statistical mean) should be equal to the scheduled period of the real-time task, and that it should be symmetrical with most of the samples focused on the center. To ensure that these requirements are met, a data adjustment process, which omits any outliers in the expense of accuracy, is presented. Then, the Z-score of the distribution according to the user-defined deviation limit provides a probability which determines the periodicity of the real-time task. Experiments are conducted to analyze the timing measurements of real-time tasks based on real-time Linux extensions of Xenomai and RT-Preempt. The results indicate that the proposed method is able to provide easier interpretation of the periodicity of real-time tasks which are valuable especially in comparing the performance of various real-time systems.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.4
• /
• pp.64-71
• /
• 2015

In this paper, we present a multi-period 0-1 knapsack problem which has the cardinality constraints. Theoretically, the presented problem can be regarded as an extension of the multi-period 0-1 knapsack problem. In the multi-period 0-1 knapsack problem, there are n jobs to be performed during m periods. Each job has the execution time and its completion gives profit. All the n jobs are partitioned into m periods, and the jobs belong to i-th period may be performed not later than in the i-th period, i = 1, ${\cdots}$, m. The total production time for periods from 1 to i is given by $b_i$ for each i = 1, ${\cdots}$, m, and the objective is to maximize the total profit. In the extended problem, we can select a specified number of jobs from each of periods associated with the corresponding cardinality constraints. As the extended problem is NP-hard, the branch and bound method is preferable to solve it, and therefore it is important to have efficient procedures for solving its linear programming relaxed problem. So we intensively explore the LP relaxed problem and suggest a polynomial time algorithm. We first decompose the LP relaxed problem into m subproblems associated with each cardinality constraints. Then we identify some new properties based on the parametric analysis. Finally by exploiting the special structure of the LP relaxed problem, we develop an efficient algorithm for the LP relaxed problem. The developed algorithm has a worst case computational complexity of order max[$O(n^2logn)$, $O(mn^2)$] where m is the number of periods and n is the total number of jobs. We illustrate a numerical example.