REACTION 2012. First International Workshop on Real-time and distributed computing in emerging applications

Permanent URI for this collection

http://dspace.uc3m.es/handle/10016/40132

Archivo Abierto Institucional de la Universidad Carlos III de Madrid: REACTION 2012. First International Workshop on Real-time and distributed computing in emerging applications

Preliminary pages

Message from the General Chairs

Keynote talk

Challenges in Distributed Real-Time Systems: Scheduling and Synchronization.Binoy Ravindran

Scheduling and middleware for emerging applications and domains

A Schedulability Analysis Framework for Real-Time Infrastructure Systems Managing Heterogeneous Resources. Jinhyun Kim and Sungwon Kang
Towards a reconfiguration service for distributed real-time Java. Pablo Basanta Val and Marisol García Valls
A dynamic service lookup and discovery scheme using a self organizing overlay network for indoor location-based service. Seol Yooung Jeong, Hyeong Gon Jo, and Soon Ju Kang
A practical solution for functional reconfiguration of real-time service based applications through partial schedulability. Marisol García Valls and Pablo Basanta Val

Cloud computing and cyber physical systems

Towards a generic, service oriented framework for distributed real-time systems. M. Tariq, M. Al Faruque, S. Grijalva, and Marilyn Wolf
Architecture of a cyberphysical avatar. Song Han, Aloysius K. Mok, Jianyong Meng, Yi-Hung Wei, Pei-chi Huang, Xiuming Zhu, Luis Sentis, Kwan Suk Kim, Risto Miikkulainen, and Jacob Menashe
A Framework for Effective Placement of Virtual Machine Replicas for Highly Available Performance-sensitive Cloud-based Applications. Kyoungho An, Faruk Caglar, Shashank Shekhar, and Anirrudha Gokhale

Modeling and traffic simulation

Toward GPU-accelerated traffic simulation and its real-time challenge.Manato Hirabayashi, Shinpei Kato, Masato Edahiro, and Yuki Sugiyama
Real-time modelling of DDS for event-driven applications. Héctor Pérez Tijeiro and J. Javier Gutiérrez García
Supporting Early Modeling and End-to-End Timing Analysis of Vehicular Distributed Real-Time Applications.Saad Mubeen, Jukka Maki-Turja, and Mikael Sjodin

Browse

Now showing 1 - 13 of 13

Message from the General Chairs
(Universidad Carlos III de Madrid, 2012-12-04) García Valls, María Soledad; Cucinotta, Tommaso; García Valls, Marisol; Cucinotta, Tommaso
Preliminary pages
(Universidad Carlos III de Madrid, 2012-12-04) García Valls, Marisol; Cucinotta, Tommaso
Towards A Generic, Service-Oriented Framework for Distributed Real-Time Systems
(Universidad Carlos III de Madrid, 2012-12-04) Tariq, Muhammad Umer; Al Faruque, Mohammad Abdullah; Grijalva, Santiago; Wolf, Marilyn; García Valls, Marisol; Cucinotta, Tommaso
Continuously increasing complexity and scale of distributed real-time systems have exposed the limitations of their existing development methodologies. This fact is evident by the unsustainable rate of increase in the development and maintenance costs of such systems. In this paper, we present a generic, service-oriented framework for distributed real-time systems. The proposed framework can potentially serve as the basis for a widely applicable, cross-domain toolset, thus, decreasing the development and maintenance costs for distributed real-time systems. The proposed framework consists of a generic, service-oriented deployment platform that abstracts away the details of implementation platform and an associated development methodology. The proposed framework makes extensive use of the existing service-oriented technologies such as Web Services. However, it also extends these technologies for application to distributed real-time systems by introducing QoS-aware service deployment and service monitoring phases. This paper presents the details of the proposed framework as well as a case-study of the application of the proposed framework to the domain of smart grid
Challenges in Distributed Real-Time Systems: Scheduling and Synchronization
(Universidad Carlos III de Madrid, 2012-12-04) Ravindran, Binoy; García Valls, Marisol; Cucinotta, Tommaso
Towards a Reconfiguration Service for Distributed Real-Time Java
(Universidad Carlos III de Madrid, 2012-12-04) Basanta Val, Pablo; García Valls, María Soledad; García Valls, Marisol; Cucinotta, Tommaso
Ancient monolithic distributed systems were attached to well-known development practices and offline analysis. Current scenarios are more dynamic, and open, plenty of applications and services which appear and disappear dynamically at runtime. Likewise, these scenarios require taking into account actions that were traditionally addressed offline, this time in an online scenario. This paper contributes a reconfiguration service in the context of distributed real-time Java application as a means to include real-time reconfiguration into next generation real-time Java systems. The paper addresses the integration taking into account changes required in the API and the cost of some reconfiguration strategies.
A Schedulability Analysis Framework for Real-time Infrastructure Systems Managing Heterogeneous Resources
(Universidad Carlos III de Madrid, 2012-12-04) Kim, Jinhyun; Kang, Sungwon; García Valls, Marisol; Cucinotta, Tommaso
Electricity generating systems, such as smart grid systems, and water management systems are infrastructure systems that manage resources critical to human life. In the systems, resources are produced and managed to supply them to various consumers, such as building, car, factory, and household, according to their needs and priorities. Reliable supply of resources depends not only on sufficient production of resources but also on reliable sharing of resource supply facilities. This paper presents a schedulability analysis framework. A prominent characteristic of the framework is that it considers at once the two types of resources, i.e. consumable resources, such as electricity, energy, and water, and sharable resources, such as pipelines, storages, and processors, are considered. To apply a formal approach to schedulability analysis of infrastructure system, this paper classifies the types of resources and real-time jobs for infrastructure systems. Then based on the classification , it presents an architectural model and a schedulability analysis framework.
A practical solution for functional reconfiguration of real-time service based applications through partial schedulability
(Universidad Carlos III de Madrid, 2012-12-04) García Valls, María Soledad; Basanta Val, Pablo; García Valls, Marisol; Cucinotta, Tommaso
Timely reconfiguration in distributed real-time systems is a complex problem with many sides to it ranging from system-wide concerns down to the intrinsic non-robust nature of the specific middleware software and the used programming techniques. In an completely open distributed system, it is not possible to achieve time-deterministic functional reconfiguration; the set of possible target configurations that the system can transition to could be extremely large threatening the temporal predictability of the reconfiguration process. Therefore, a set of bounds and limitations to the structure of systems and to their open nature need to be imposed. In this paper, we present the different sides of the problem of reconfiguration. We provide a solution for timely reconfiguration based on reducing the solution space of solutions of partially closed applications; we have enhanced the logic of a middleware for distributed soft real-time applications with the proposed technique. As a result, applications require a limited number of schedulability tests to search for the valid target configuration. We present some results on the actual reduction of the configuration space achieved by our middleware.
A Framework for Effective Placement of Virtual Machine Replicas for Highly Available Performance-sensitive Cloud-based Applications
(Universidad Carlos III de Madrid, 2012-12-04) An, Kyoungho; Caglar, Faruk; Shekhar, Shashank; Gokhale, Aniruddha; García Valls, Marisol; Cucinotta, Tommaso
Applications are increasingly being deployed in the Cloud due to benefits stemming from economy of scale, scalability, flexibility and utility-based pricing model. Although most cloud-based applications have hitherto been enterprisestyle, there is a new trend towards hosting performancesensitive applications in the cloud that demand both high availability and good response times. In the current stateof- the-art in cloud computing research, there does not exist solutions that provide both high availability and acceptable response times to these applications in a way that also optimizes resource consumption in data centers, which is a key consideration for cloud providers. This paper addresses this dual challenge by presenting a design of a fault-tolerant framework for virtualized data centers that makes two important contributions. First, it describes an architecture of a fault-tolerance framework that can be used to automatically deploy replicas of virtual machines in data centers in a way that optimizes resources while assures availability and responsiveness. Second, it describes a specific formulation of a replica deployment combinatorial optimization problem that can be plugged into our strategizable deployment framework.
Architecture of a Cyberphysical Avatar
(Universidad Carlos III de Madrid, 2012-12-04) Han, Song; Mok, Aloysius K.; Meng, Jianyong; Wei, Yi-Hung; Huang, Pei-Chi; Zhu, Xiuming; Sentis, Luis; García Valls, Marisol; Cucinotta, Tommaso
This paper introduces the concept of a cyberphysical avatar which is defined to be a semi-autonomous robotic system that adjusts to an unstructured environment and performs physical tasks subject to critical timing constraints while under human supervision. Cyberphysical avatar integrates the recent advance in three technologies: body-compliant control in robotics, neuroevolution in machine learning and QoS guarantees in realtime communication. Body-compliant control is essential for operator safety since cyberphysical avatars perform cooperative tasks in close proximity to humans. Neuroevolution technique is essential for ”programming” cyberphysical avatars inasmuch as they are to be used by non-experts for a large array of tasks, some unforeseen, in an unstructured environment. QoS-guaranteed realtime communication is essential to provide predictable, boundedtime response in human-avatar interaction. By integrating these technologies, we have built a prototype cyberphysical avatar testbed.
A Dynamic Service Lookup and Discovery Scheme using a Self-Organizing Overlay Network for Indoor Location-Based Service
(Universidad Carlos III de Madrid, 2012-12-04) Jeong, Seol Young; Jo, Hyeong Gon; Kang, Soon Ju; García Valls, Marisol; Cucinotta, Tommaso
Indoor location-based service (LBS) has various challenges, including that of numerous service lookups being requested concurrently and frequently from several locations, and that the network infrastructure needs to support high scalability, such as inserting or deleting network nodes anytime and anywhere. In general, indoor LBS resources are generally located in close proximity to the requested point. However, a traditional centralized LBS system needs to maintain a geographical map of the entire building or complex in its central server, which can cause low scalability and traffic congestion. This paper presents a self-organizing and fully distributed indoor LBS platform through regional cooperation among devices, and a service lookup algorithm that searches for the shortest physical path to the service resource. An evaluation of the performance of the proposed platform has been compared to the traditional centralized method regarding the service turnaround time according to the number of concurrent lookup increases.
Supporting Early Modeling and End-to-end Timing Analysis of Vehicular Distributed Real-Time Applications
(Universidad Carlos III de Madrid, 2012-12-04) Mubeen, Saad; Sjödin, Mikael; Mäki-Turja, Jukka; García Valls, Marisol; Cucinotta, Tommaso
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.
Real-time modelling of DDS for event-driven applications
(Universidad Carlos III de Madrid, 2012-12-04) Pérez Tijeiro, Héctor; Gutiérrez, J. Javier; García Valls, Marisol; Cucinotta, Tommaso
The Data Distribution Service (DDS) standard defines a data-centric distribution middleware that supports the development of distributed real-time systems. To this end, the standard includes a wide set of configurable parameters to provide different degrees of Quality of Service (QoS). This paper presents an analysis of these QoS parameters when DDS is used to build reactive applications normally designed under an event-driven paradigm, and shows how to configure DDS to obtain predictable applications suitable to apply traditional schedulability analysis techniques.
Toward GPU-accelerated Traffic Simulation and Its Real-Time Challenge
(Universidad Carlos III de Madrid, 2012-12-04) Hirabayashi, Manato; Kato, Shinpei; Edahiro, Masato; Sugiyama, Yuki; García Valls, Marisol; Cucinotta, Tommaso
Traffic simulation is a growing domain of computational physics. Many life and industrial applications would benefit from traffic simulation to establish reliable transportation systems. A core challenge of this science research, however, is its unbounded scale of computation. This paper explores an advantage of using the graphics processing unit (GPU) for this computational challenge. We study two schemes of maximizing GPU performance in the context of traffic simulation, and provide some basic experiments. The experimental results show that our GPU implementation improves simulation speed by five times over the traditional CPU implementation. We also discuss that additional orders-ofmagnitude improvements could be achieved by overcoming the current hardware limitation of the GPU.

Browse

Recent Submissions