Search

End-to-end data-propagation delay analysis allows verification of important timing constraints, such as age and reaction, that areoften specified on chains of tasks and messages in real-time systems.We identify that the existing analysis does not support distributed taskchains that include the Time-Sensitive Networking (TSN) messages. Tothis end, this paper extends the existing analysis to allow the end-to-endtiming analysis of distributed task chains that include TSN messages.The extended analysis supports all types of traffic in TSN, includingthe Scheduled Traffic (ST), Audio Video Bridging (AVB), and BestEffort (BE) traffic. Furthermore, the extended analysis accounts for thesynchronization among the end stations that are connected via TSN.The applicability of the analysis is demonstrated using an automotiveapplication case study.

Data movement has long been identified as the biggest challenge facing modern computer systems' designers. To tackle this challenge, many novel data compression algorithms have been developed. Often variable rate compression algorithms are favored over fixed rate. However, variable rate decompression is difficult to parallelize. Most existing algorithms adopt a single parallelization strategy suited for a particular HW platform. Such an approach fails to harness the parallelism found in diverse modern HW architectures. We propose a parallelization method for tiled variable rate compression algorithms that consists of multiple strategies that can be applied interchangeably. This allows an algorithm to apply the strategy most suitable for a specific HW platform. Our strategies are based on generating metadata during encoding, which is used to parallelize the decoding process. To demonstrate the effectiveness of our strategies, we implement them in a state-of-the-art compression algorithm called ZFP. We show that the strategies suited for multicore CPUs are different from the ones suited for GPUs. On a CPU, we achieve a near optimal decoding speedup and an overhead size which is consistently less than 0.04% of the compressed data size. On a GPU, we achieve average decoding rates of up to 100 GiB/s. Our strategies allow the user to make a trade-off between decoding throughput and metadata size overhead.

With an increasing amount of textual information available there is also an increased need to make this information more accessible. Our paper describes a modified TextRank model and investigates the different methods available to use automatic text summarization as a means for summary creation of swedish news articles. To evaluate our model we focused on intrinsic evaluation methods, in part through content evaluation in the form of of measuring referential clarity and non-redundancy, and in part by text quality evaluation measures, in the form of keyword retention and ROUGE evaluation. The results acquired indicate that stemming and improved stop word capabilities can have a positive effect on the ROUGE scores. The addition of redundancy checks also seems to have a positive effect on avoiding repetition of information. Keyword retention decreased somewhat, however. Lastly all methods had some trouble with dangling anaphora, showing a need for further work within anaphora resolution.

Providing computer-based services for vehicle functions has evolved to the point where a large majority of functions are realized by software. However, the need to provide safety and security in critical functions such as braking, steering, motor control, etc. requires an approach that can guarantee the continuous reliable operation of the functions. At the same time, there are a variety of functions that are less critical from the vehicle operation perspective that can be provided where safety and security are less critical. From a vehicle manufacturers point of view, providing both types of services in an economic and reliable manner is a real challenge. To meet this challenge, we consider the Rubus Tool Suit for the software development and a well-proven (in industrial use for over twenty years) and certified (according to ISO 26262) operating system Kernel for its execution. In addition, a user-friendly approach to model-and component-based development concept called the software circuits has provided an approach to meet the demands of both safety-critical deterministic and as well as non-safety critical non-deterministic services. In this paper, a brief history of the evolution of Rubus approach as well as an overview of the driving concepts used in providing the Rubus products are described.

In this paper we present a success story of academic-industrial collaboration in the vehicular domain. The collaboration has resulted in the development and evolution of a model-and component-based software development tool chain that is used to develop safety-critical, robust, and certified, control software for vehicles. The tool chain has been successfully used in the vehicle industry for about 20 years. The sustained development of the tool chain is based one a unique collaboration, described in this paper, where the collaborators form a clear value chain from academia, through tool-developer, to the end-users of the technology. We describe experiences of each collaborator with a focus on the end-user's perspective. Moreover, we highlight some ongoing and future works within this collaboration.

Dynamic RAM (DRAM) is a source of memory contention and interference problems on commercial of the shelf (COTS) multicore architectures. Due to its variable access time, it can greatly influence the task's WCET and can lead to unpredictability. In this paper, we provide a worst case delay analysis for a DRAM memory request to safely bound memory contention on multicore architectures. We derive a worst-case service time for a single memory request and then combine it with the per-request memory interference that can be generated by the tasks executing on same or different cores in order to generate the delay bound.

The existing tools for the response-time analysis of Controller Area Network (CAN) support only periodic and sporadic messages. They do not analyze mixed messages which are implemented by several higher-level protocols based on CAN that are used in the automotive industry. We present a new response-time analyzer for CAN that supports periodic and sporadic as well as mixed messages. Moreover, it supports the analysis of the system where periodic and mixed messages are scheduled with offsets. It will support the analysis of all types of messages while taking into account several queueing policies and buffer limitations in the CAN controllers.

The existing offset-aware response-time analysis of Controller Area Network (CAN) for mixed messages has certain practical limitations. It is based on the assumption that the jitter and deadline of a message are smaller or equal to the transmission period. However, practical systems may contain messages with release jitter greater than the period. Consequently, the deadlines specified for such messages are also greater than their periods. In this paper, we extend the existing response-time analysis for mixed messages in CAN that are scheduled with offsets and have arbitrary jitter and deadlines. Mixed messages are implemented by several higher-level protocols based on CAN that are used in the automotive industry. The extended analysis is applicable to any higher-level protocol for CAN that uses periodic, sporadic and mixed transmission modes.

This paper presents two strategies to support communication between real-time executable Runnable Virtual Node (RVN) components in the ProCom component technology. We describe the currently implemented server-based communication strategy which uses a dedicated server for communication. We compare the server-based technique with a direct (RVN-to-RVN) communication strategy. The paper also describes how these strategies could be evaluated for real-time performance, and the real-time analysis technologies needed to perform such an evaluation.

REACTION 2012. 1st International workshop on Real-time and distributed computing in emerging applications. December 4th, 2012, San Juan, Puerto Rico. The current model- and component-based development approaches for automotive distributed real-time systems have non-existing, or limited, support for modeling network traffic originating from outside the vehicle, i.e., vehicle-tovehicle, vehicle-to-infrastructure, and cloud-based applications. We present novel modeling and analysis techniques to allow early end-to-end timing analysis of distributed applications based on their models and simple models of network traffic that originates from outside of the model. As a proof of concept, we implement these techniques in the existing industrial tool suite Rubus- ICE which is used for the development of software for vehicular embedded systems by several international companies. We also conduct an application-case study to validate our techniques. This work is supported by the Swedish Knowledge Foundation (KKS) within the project FEMMVA. We thank the industrial partners Arcticus Systems, BAE Systems Hägglunds and Volvo Construction Equipment (VCE), Sweden.

This paper presents extensions to the previous implementationof two-level Hierarchical Scheduling Framework(HSF) for FreeRTOS. The results presented here allow the useof HSF for FreeRTOS in hard-real time applications, with thepossibility to include legacy applications and components notexplicitly developed for hard real-time or the HSF. Specifically, we present the implementations of (i) global andlocal resource sharing using the Hierarchical Stack ResourcePolicy and Stack Resource Policy respectively, (ii) kernel supportfor the periodic task model, and (iii) mapping of original FreeRTOSAPI to the extended FreeRTOS HSF API. We also presentevaluations of overheads and behavior for different alternativeimplementations of HSRP with overrun from experiments on theAVR 32-bit board EVK1100. In addition, real-time schedulinganalysis with models of the overheads of our implementation ispresented.

We present the concept of a virtual node as means to achieve predictable integration of software components with real-time requirements. The virtual node is based on the technology using the hierarchical scheduling framework to achieve temporal isolation between components (or sets of components). The temporal isolation means that all components deployed to a virtual node will retain their temporal properties when integrated with other components on a physical node. We present how the concept of virtual nodes is applicable to three different component technologies: ProCom, Autosar and AADL. Keywords-real-time systems; component integration