Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365652
C. Cardeira, Z. Mammeri
We analyse the use of artificial neural networks (ANNs) to approximate solving scheduling problems. It is well established that the ANNs main advantage is the small amount of time they take to find an approximate solution, but a question arises: what about the optimality of the obtained solution? A considerable variety of work has been carried out on this subject but, unfortunately, the majority of the studies have focused on the analysis of the classical TSP problem. The obtained results are useful as a reference but can't be directly extrapolated for real-time systems. We analyse the behaviour of an ANN based scheduling algorithm when scheduling tasks in a real-time system, using the baseline task set from the Hartstone Benchmark which is considered as a typical set for some real-time applications.<>
{"title":"Performance analysis of a neural network based scheduling algorithm","authors":"C. Cardeira, Z. Mammeri","doi":"10.1109/WPDRTS.1994.365652","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365652","url":null,"abstract":"We analyse the use of artificial neural networks (ANNs) to approximate solving scheduling problems. It is well established that the ANNs main advantage is the small amount of time they take to find an approximate solution, but a question arises: what about the optimality of the obtained solution? A considerable variety of work has been carried out on this subject but, unfortunately, the majority of the studies have focused on the analysis of the classical TSP problem. The obtained results are useful as a reference but can't be directly extrapolated for real-time systems. We analyse the behaviour of an ANN based scheduling algorithm when scheduling tasks in a real-time system, using the baseline task set from the Hartstone Benchmark which is considered as a typical set for some real-time applications.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132396956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365628
H. Schepers
A compositional network proof theory to specify and verify fault tolerant real-time distributed systems is presented. In this theory the failure hypothesis of a system is formalized as a relation between the system's normal behaviour (i.e., the behaviour that conforms to the specification) and its acceptable, behaviour, that is, the normal behaviour together with the exceptional behaviour (i.e., the behaviour whose abnormality should be tolerated).<>
{"title":"Specifying and verifying fault tolerant real-time distributed systems compositionally","authors":"H. Schepers","doi":"10.1109/WPDRTS.1994.365628","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365628","url":null,"abstract":"A compositional network proof theory to specify and verify fault tolerant real-time distributed systems is presented. In this theory the failure hypothesis of a system is formalized as a relation between the system's normal behaviour (i.e., the behaviour that conforms to the specification) and its acceptable, behaviour, that is, the normal behaviour together with the exceptional behaviour (i.e., the behaviour whose abnormality should be tolerated).<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132447552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365648
Young-Kuk Kim, M. Lehr, D. George, S. Son
The design and implementation of database systems for real-time applications presents many new and challenging problems. In addition to maintaining data consistency, the database system must satisfy timing constraints associated with transactions. In this paper we present our experiences in integrating a database server with a real-time operating system kernel and our attempts at providing flexible control for concurrent transaction management. We chose the ARTS operating system kernel as the basis for the real-time database server. Current research issues involving the development of a programming interface and imprecise computing server are also discussed.<>
{"title":"A database server for distributed real-time systems: issues and experiences","authors":"Young-Kuk Kim, M. Lehr, D. George, S. Son","doi":"10.1109/WPDRTS.1994.365648","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365648","url":null,"abstract":"The design and implementation of database systems for real-time applications presents many new and challenging problems. In addition to maintaining data consistency, the database system must satisfy timing constraints associated with transactions. In this paper we present our experiences in integrating a database server with a real-time operating system kernel and our attempts at providing flexible control for concurrent transaction management. We chose the ARTS operating system kernel as the basis for the real-time database server. Current research issues involving the development of a programming interface and imprecise computing server are also discussed.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117172868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365629
M. Mutka
The demand for systems with predictable communication delays is growing as the number of applications for real-time systems expands. We present a design for guaranteeing deadline constraints for real-time communications in a wormhole network. An important feature of the design is that it requires few changes to current wormhole networks that support logical channels. The design approach applies rate monotonic scheduling technology to wormhole networks, which allows us to verify whether an assignment of real-time traffic can have deadlines guaranteed. We provide a tool that analyzes the feasibility of real-time traffic assignments, which aids designers who develop applications that use real-time communication features of a wormhole network.<>
{"title":"Using rate monotonic scheduling technology for real-time communications in a wormhole network","authors":"M. Mutka","doi":"10.1109/WPDRTS.1994.365629","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365629","url":null,"abstract":"The demand for systems with predictable communication delays is growing as the number of applications for real-time systems expands. We present a design for guaranteeing deadline constraints for real-time communications in a wormhole network. An important feature of the design is that it requires few changes to current wormhole networks that support logical channels. The design approach applies rate monotonic scheduling technology to wormhole networks, which allows us to verify whether an assignment of real-time traffic can have deadlines guaranteed. We provide a tool that analyzes the feasibility of real-time traffic assignments, which aids designers who develop applications that use real-time communication features of a wormhole network.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133465726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365635
N. Karangelen, Trident
This paper summarizes several existing techniques for representing system behavior and describes the key features of an advanced approach for capturing behavior for a large, complex, real-time, computer-based system within a multi-domain system design capture and analysis methodology. Within this advanced approach for capturing system behavior three key aspects of representing complex system behavior are addressed: (I) response of the system to stimuli as observed externally (i.e., system as a black box), (2) real-time behavior of the system logical design options (i.e. Control, timing, and synchronization aspects of the system functions and data without regard to specific implementation in resources), and (3) real-time behavior of system implementation design options (i.e. response time, delays, queue lengths, and other aspects of the system behavior as embodied in the hardware, software and human resources and resource architectures).<>
{"title":"Representing system behavior in design and analysis of large complex real-time systems","authors":"N. Karangelen, Trident","doi":"10.1109/WPDRTS.1994.365635","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365635","url":null,"abstract":"This paper summarizes several existing techniques for representing system behavior and describes the key features of an advanced approach for capturing behavior for a large, complex, real-time, computer-based system within a multi-domain system design capture and analysis methodology. Within this advanced approach for capturing system behavior three key aspects of representing complex system behavior are addressed: (I) response of the system to stimuli as observed externally (i.e., system as a black box), (2) real-time behavior of the system logical design options (i.e. Control, timing, and synchronization aspects of the system functions and data without regard to specific implementation in resources), and (3) real-time behavior of system implementation design options (i.e. response time, delays, queue lengths, and other aspects of the system behavior as embodied in the hardware, software and human resources and resource architectures).<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126519938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365627
G. Bruno, A. Castella, R. Agarwal, I. Pavesio, M. P. Pescarmona
The development and maintenance of a large software project requires a coherent set of tools and methodologies in order to produce consistent specifications, documentation and code. The application of Petri nets to the modeling of large real world systems has been, up to recent years, hindered by the lack of effective mechanisms for structuring and analyzing net based models in a modular way. On the other hand, the introduction of hierarchical CPN's and of other analogous extensions of High-level Petri nets allows the modeled system to be inspected at varying levels of detail and facilitates re-use of components of the model. This paper presents the evaluation of some of these new methodologies for the development of a real project and the effective use of one of them.<>
{"title":"Using modular Petri nets for developing telecommunication software","authors":"G. Bruno, A. Castella, R. Agarwal, I. Pavesio, M. P. Pescarmona","doi":"10.1109/WPDRTS.1994.365627","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365627","url":null,"abstract":"The development and maintenance of a large software project requires a coherent set of tools and methodologies in order to produce consistent specifications, documentation and code. The application of Petri nets to the modeling of large real world systems has been, up to recent years, hindered by the lack of effective mechanisms for structuring and analyzing net based models in a modular way. On the other hand, the introduction of hierarchical CPN's and of other analogous extensions of High-level Petri nets allows the modeled system to be inspected at varying levels of detail and facilitates re-use of components of the model. This paper presents the evaluation of some of these new methodologies for the development of a real project and the effective use of one of them.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128461128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365649
S. Ronngren, B. Shirazi, D. Lorts
Suboptimal solutions to the NP-complete problem of task scheduling in a multiprocessing system are achievable with the aid of heuristic methods. Static scheduling heuristics for real-time multiprocessing systems are typically based on existing algorithms developed for non-real-time systems. Unfortunately this approach results in the real-rime heuristics inheriting the deficiencies of the non-real-time algorithms as well. Existing scheduling heuristics compromise the results of the scheduling effort by insufficiently representing the task characteristics of an application graph. In this paper we present results of experimentation in which the parameters of the DAG are enhanced to more accurately correspond to those of real-world real-time applications. A method of specifying weighted combinations and priorities of simple scheduling heuristics as the scheduling algorithm is presented. Results of the compound heuristics are compared to the results of previous work in the field with some interesting conclusions.<>
{"title":"Empirical evaluation of weighted and prioritized static scheduling heuristics for real-time multiprocessing","authors":"S. Ronngren, B. Shirazi, D. Lorts","doi":"10.1109/WPDRTS.1994.365649","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365649","url":null,"abstract":"Suboptimal solutions to the NP-complete problem of task scheduling in a multiprocessing system are achievable with the aid of heuristic methods. Static scheduling heuristics for real-time multiprocessing systems are typically based on existing algorithms developed for non-real-time systems. Unfortunately this approach results in the real-rime heuristics inheriting the deficiencies of the non-real-time algorithms as well. Existing scheduling heuristics compromise the results of the scheduling effort by insufficiently representing the task characteristics of an application graph. In this paper we present results of experimentation in which the parameters of the DAG are enhanced to more accurately correspond to those of real-world real-time applications. A method of specifying weighted combinations and priorities of simple scheduling heuristics as the scheduling algorithm is presented. Results of the compound heuristics are compared to the results of previous work in the field with some interesting conclusions.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124705160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365655
Jyh-Charn S. Liu, Junguk L. Kim, Soojung Lee, T. Urbanik
Ramp metering is an important technique to effectively utilize the freeway capacity by regulating the traffic flow of freeway entrance ramps. In this extended abstract, we discuss the outline of the free flow control algorithm of the Advanced distributed Ramp Metering System, ARMS, for the real-time control of freeway ramp metering systems. ARMS consists of three levels of control algorithms integrated together for free flow control, breakdown prediction, and congestion resolution. Based on a free flow traffic model, the free flow control algorithm takes into account of the congestion risk when it tries to maximize the traffic throughput. The breakdown prediction algorithm can make prediction on the possibility of imminent congestion caused by flow breakdown, and the congestion resolution algorithm is to resolve the congestion if it actually occurs. The ARMS has a modular architecture so that it can be scaled for incremental implementation. To control freeway traffic, a system wide optimization objective is first defined for a target area. Then, ramp metering decisions are made by the entrance ramp controllers within the target area after their traffic data are exchanged.<>
{"title":"The advanced distributed ramp metering system (ARMS)","authors":"Jyh-Charn S. Liu, Junguk L. Kim, Soojung Lee, T. Urbanik","doi":"10.1109/WPDRTS.1994.365655","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365655","url":null,"abstract":"Ramp metering is an important technique to effectively utilize the freeway capacity by regulating the traffic flow of freeway entrance ramps. In this extended abstract, we discuss the outline of the free flow control algorithm of the Advanced distributed Ramp Metering System, ARMS, for the real-time control of freeway ramp metering systems. ARMS consists of three levels of control algorithms integrated together for free flow control, breakdown prediction, and congestion resolution. Based on a free flow traffic model, the free flow control algorithm takes into account of the congestion risk when it tries to maximize the traffic throughput. The breakdown prediction algorithm can make prediction on the possibility of imminent congestion caused by flow breakdown, and the congestion resolution algorithm is to resolve the congestion if it actually occurs. The ARMS has a modular architecture so that it can be scaled for incremental implementation. To control freeway traffic, a system wide optimization objective is first defined for a target area. Then, ramp metering decisions are made by the entrance ramp controllers within the target area after their traffic data are exchanged.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116826450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365650
E. Haddad
The problem of optimizing the balancing of processing load originating at the various sites of a bus-interconnected set of heterogeneous processors is examined. The distributed load is assumed to be arbitrarily divisible with no particular affinity to any of the processing sites. Minimization of processing completion time is the objective of load redistribution via the limited communication capacity of the bus. The optimal amounts of load exchange among the sending and receiving processors are derived. Load transfer across the bus proceeds concurrently with load execution at the various sites. Absolute optimality can be achieved only if synchronization delay is eliminated through proper real-time delivery of communicated load. The optimal load transmission and delivery time-schedule is formulated in terms of the load transfer rates pertaining to each sending and receiving site. The available communication rate of the bus is partitioned among the site rates, which are found to be piece-wise constant functions of time.<>
{"title":"Real-time optimization of distributed load balancing","authors":"E. Haddad","doi":"10.1109/WPDRTS.1994.365650","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365650","url":null,"abstract":"The problem of optimizing the balancing of processing load originating at the various sites of a bus-interconnected set of heterogeneous processors is examined. The distributed load is assumed to be arbitrarily divisible with no particular affinity to any of the processing sites. Minimization of processing completion time is the objective of load redistribution via the limited communication capacity of the bus. The optimal amounts of load exchange among the sending and receiving processors are derived. Load transfer across the bus proceeds concurrently with load execution at the various sites. Absolute optimality can be achieved only if synchronization delay is eliminated through proper real-time delivery of communicated load. The optimal load transmission and delivery time-schedule is formulated in terms of the load transfer rates pertaining to each sending and receiving site. The available communication rate of the bus is partitioned among the site rates, which are found to be piece-wise constant functions of time.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114326121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-04-28DOI: 10.1109/WPDRTS.1994.365632
A. Esposito, R. Vaccaro, L. Verdoscia
Parallel systems usually increase the complexity of evaluating the program execution time, so as to verify timing constraints into real-time environments. For this reason, the approach described in this paper is based on a massively parallel architecture using VLSI to implement in a fine-grain mode the data-flow paradigm. The system employs the static data-flow execution model that does not need to use a control token. A cluster of functional units has been simulated in order to validate the proposed architecture and the first results confirm that the overall behaviour is predictable.<>
{"title":"A special-purpose parallel system for predictable real-time systems","authors":"A. Esposito, R. Vaccaro, L. Verdoscia","doi":"10.1109/WPDRTS.1994.365632","DOIUrl":"https://doi.org/10.1109/WPDRTS.1994.365632","url":null,"abstract":"Parallel systems usually increase the complexity of evaluating the program execution time, so as to verify timing constraints into real-time environments. For this reason, the approach described in this paper is based on a massively parallel architecture using VLSI to implement in a fine-grain mode the data-flow paradigm. The system employs the static data-flow execution model that does not need to use a control token. A cluster of functional units has been simulated in order to validate the proposed architecture and the first results confirm that the overall behaviour is predictable.<<ETX>>","PeriodicalId":275053,"journal":{"name":"Second Workshop on Parallel and Distributed Real-Time Systems","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133201269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：