Meeting Real-Time Requirements with Multi-core Processors

Kastner, Daniel L.; Schlickling, Marc; Pister, Markus; Cullmann, Christoph; Gebhard, Gernot; Heckmann, Reinhold; Ferdinand, Christian

doi:10.1007/978-3-642-33675-1_10

Cited by 11 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…7,21 As an alternative to tilting at windmills, that is, trying to fix designs that are less favorable for WCET analysis, one can try to design the architectures appropriately in the first place. 42,44 Appropriately meaning that one can easily predict their behavior while they still exhibit high performance.…”

Section: Academic Developmentmentioning

confidence: 99%

Computation takes time, but how much?

Wilhelm

Grund

2014

Commun. ACM

View full text Add to dashboard Cite

A few yeArs ago, in the pages of this magazine, Edward Lee argued that computing needs time. 23 This article focuses on the natural assumption that computing also takes time. We examine the problem of determining how much time. It is the problem of verifying the real-time behavior of safety-critical embedded systems. For such systems, for example, anti-lock brakes and airbags, punctual behavior is of utmost importance: If the controlling computations take too long, quality of service degrades or the systems fail completely-your braking distance is longer or your head hits the steering wheel, respectively.The basis for verifying the timeliness of system reactions is reliable information on the execution times of all computational tasks involved. It is the job of timing analysis, also called worst-case execution-time (WCET) analysis, to determine such information. key insights the tremendous progress in microprocessor architecture not only increased average-case performance, but also the complexity of verifying the real-time behavior of programs executed on those architectures.to derive useful execution time guarantees, static analyses must prove that speculation mechanisms of modern cPus will indeed be effective during program runtime. this is challenging due to timing anomalies and interdependencies between architectural components. computation takes time, But how much?

show abstract

Section: Academic Developmentmentioning

confidence: 99%

Computation takes time, but how much?

Wilhelm

Grund

2014

Commun. ACM

View full text Add to dashboard Cite

show abstract

“…In general, the research on WCET estimation on multi-core architectures has mainly focused on the predictability of accesses to shared resources. For example, the predictability features and the timing variance of single and multi-core processors are analyzed in [6]. Particularly, the research on high-level flow analysis of parallel applications running on multi-core processors is relatively new and very few studies are available to compare.…”

Section: Related Workmentioning

confidence: 99%

Timing Analysis of Parallel Embedded Real-Time Software–practical Experiences

2016

Proceedings of 2016 the 6th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

The knowledge about the Worst-Case Execution Time (WCET) of a program is fundamental to the successful design of embedded real-time software. The WCET analysis of embedded software is also needed for tasks scheduling, understanding the system and to guarantee its behavior. The conventional WCET analysis deals with the sequential code running on uni-processors. However, the increasing use of multi-core processors in embedded real-time systems; the WCET analyses face new problems. This article reports our experiences of performing the WCET analysis of Parallel Embedded Software (PES) running on multi-core hardware. The main objective was to investigate the way WCET estimates of PES can be computed both statically and dynamically. The experiences reported in this article include the challenges faced, possible solutions to these challenges and the workarounds developed. This article also provides observations on the benefits and drawbacks of deriving the WCET estimates and sets path for future research in this direction.

show abstract

“…The current research efforts of WCET analysis for multicore systems are focused on performance enhancing hardware features [19,12,15], and application [16,18,4] or programming level [27,28,9]. However, there is lack of research in test-data generation for WCET estimation of parallel real-time systems executing on multicore architectures.…”

Section: Introductionmentioning

confidence: 99%

Test-Data Generation for Testing Parallel Real-Time Systems

Aziz

Shah

2015

Testing Software and Systems

View full text Add to dashboard Cite

The Worst-Case Execution Time (WCET) of real-time systems is mainly influenced by the program design, its execution environment and the input data. To cover the last factor in the context of WCET estimation, the objective of this work is to generate the test-data that maximize the execution times of the parallel real-time systems. In this paper, a test-data generation technique is proposed that uses Genetic Algorithms to automatically generate the input data, to be used for testing of parallel real-time systems. The proposed technique was applied to a parallel embedded application-Stringsearch. The result was an analysis that took as input the parallel program and generated the test-data that cause maximal execution times. The generated test-data showed improvements by exercising long execution times in comparison to randomly generated input data.

show abstract

Meeting Real-Time Requirements with Multi-core Processors

Cited by 11 publications

References 7 publications

Computation takes time, but how much?

Computation takes time, but how much?

Timing Analysis of Parallel Embedded Real-Time Software–practical Experiences

Test-Data Generation for Testing Parallel Real-Time Systems

Contact Info

Product

Resources

About