Pub Date : 1995-10-25DOI: 10.1109/RTCSA.1995.528756
H. Tezuka, T. Nakajima
Real-time system and continuous media system researchers have worked in areas that are considered by each to be unrelated to the other's domain although continuous media system researchers agree with the effectiveness of real-time technologies on continuous media systems. Media control techniques for improving the smoothness of continuous media such as jitter control and inter-stream synchronization schemes are developed for smoothing continuous media streams running on traditional timesharing operating systems. On the other hand, real-time system researchers show the effectiveness of real-time technologies for satisfying timing constraints of the continuous media. However, there is no report to showing the effectiveness of real-time technologies for the media control techniques. This paper describes experiences with building a continuous media application on Real-Time Mach. First, we present the structure of our distributed video player, called QtPlay, and its media control techniques for smoothing continuous media streams, then we show the effectiveness of real-time technologies for the media control techniques.
{"title":"Experiences with building a continuous media application on Real-Time Mach","authors":"H. Tezuka, T. Nakajima","doi":"10.1109/RTCSA.1995.528756","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528756","url":null,"abstract":"Real-time system and continuous media system researchers have worked in areas that are considered by each to be unrelated to the other's domain although continuous media system researchers agree with the effectiveness of real-time technologies on continuous media systems. Media control techniques for improving the smoothness of continuous media such as jitter control and inter-stream synchronization schemes are developed for smoothing continuous media streams running on traditional timesharing operating systems. On the other hand, real-time system researchers show the effectiveness of real-time technologies for satisfying timing constraints of the continuous media. However, there is no report to showing the effectiveness of real-time technologies for the media control techniques. This paper describes experiences with building a continuous media application on Real-Time Mach. First, we present the structure of our distributed video player, called QtPlay, and its media control techniques for smoothing continuous media streams, then we show the effectiveness of real-time technologies for the media control techniques.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115995333","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528749
P. Tsang, J. M. Ng, E. Chan, Chan-Hee Lee
This paper presents the simulation results on the FDDI medium access protocols and two modified FDDI medium access protocols. For synchronous traffic, there is usually a deadline requirement which specifies the time allowed for the packet to reach its destination. For FDDI, if the delay bound is D, the stations have to set the TTRT=D/2 in order ensure the deadline requirement can be met. This gives the total synchronous bandwidth as D/2-T/sub ring/. However, the total usable synchronous bandwidth is D-T/sub ring/ which means the capability of supporting synchronous bandwidth is halved. Zheng (1992) and Leung (1994) have proposed new schemes to overcome these problems and have no diverse effect on asynchronous throughput. Extensive simulations have been carried out on these schemes to evaluate their performance characteristics. Also, a frame-drop policy at source station is proposed and shown to be effective in improving performance on synchronous transmission.
{"title":"Performance evaluation of real-time message delivery in FDDI networks","authors":"P. Tsang, J. M. Ng, E. Chan, Chan-Hee Lee","doi":"10.1109/RTCSA.1995.528749","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528749","url":null,"abstract":"This paper presents the simulation results on the FDDI medium access protocols and two modified FDDI medium access protocols. For synchronous traffic, there is usually a deadline requirement which specifies the time allowed for the packet to reach its destination. For FDDI, if the delay bound is D, the stations have to set the TTRT=D/2 in order ensure the deadline requirement can be met. This gives the total synchronous bandwidth as D/2-T/sub ring/. However, the total usable synchronous bandwidth is D-T/sub ring/ which means the capability of supporting synchronous bandwidth is halved. Zheng (1992) and Leung (1994) have proposed new schemes to overcome these problems and have no diverse effect on asynchronous throughput. Extensive simulations have been carried out on these schemes to evaluate their performance characteristics. Also, a frame-drop policy at source station is proposed and shown to be effective in improving performance on synchronous transmission.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"185 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123208269","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528760
Stuart Shih, Young-Kuk Kim, S. Son
In conventional database systems, performance is primarily measured by the number of transactions completed within a unit time. In real-time applications, timing and criticality characteristics of transactions must be taken into account. In this paper, we examine the performance of StarBase, a firm real-time database system. The deadline guarantee ratio and average response times are the primary performance measures. There have been performance studies on real-time database systems, but most of them were performed using simulation. This work demonstrates the feasibility of developing a real-time database system with an acceptable performance.
{"title":"Performance evaluation of a firm real-time database system","authors":"Stuart Shih, Young-Kuk Kim, S. Son","doi":"10.1109/RTCSA.1995.528760","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528760","url":null,"abstract":"In conventional database systems, performance is primarily measured by the number of transactions completed within a unit time. In real-time applications, timing and criticality characteristics of transactions must be taken into account. In this paper, we examine the performance of StarBase, a firm real-time database system. The deadline guarantee ratio and average response times are the primary performance measures. There have been performance studies on real-time database systems, but most of them were performed using simulation. This work demonstrates the feasibility of developing a real-time database system with an acceptable performance.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128927927","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528773
C. Fuhrman, S. Chutani, H. Nussbaumer
A variant of N-Modular Redundancy (NMR) allows an NMR system to have up to twice the throughput while keeping the same reliability model. Each task is executed by a majority of processors. The tasks are diagnosed (by comparing their results) rather than the processors (by a separate diagnostic process). This scheme does not increase the complexity of the circuitry involved as diagnosis does not have to be adaptive. Intermittent and transient, as well as permanent faults are tolerated.
n -模块化冗余(NMR)的变体允许NMR系统在保持相同可靠性模型的情况下具有高达两倍的吞吐量。每个任务由大多数处理器执行。对任务进行诊断(通过比较它们的结果)，而不是对处理器进行诊断(通过单独的诊断过程)。这种方案不会增加电路的复杂性，因为诊断不必是自适应的。可以容忍间歇性和瞬态以及永久性故障。
{"title":"Efficient NMRCD scheme for fault tolerance in responsive systems","authors":"C. Fuhrman, S. Chutani, H. Nussbaumer","doi":"10.1109/RTCSA.1995.528773","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528773","url":null,"abstract":"A variant of N-Modular Redundancy (NMR) allows an NMR system to have up to twice the throughput while keeping the same reliability model. Each task is executed by a majority of processors. The tasks are diagnosed (by comparing their results) rather than the processors (by a separate diagnostic process). This scheme does not increase the complexity of the circuitry involved as diagnosis does not have to be adaptive. Intermittent and transient, as well as permanent faults are tolerated.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133207275","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528754
Dong-Won Park, S. Natarajan, A. Kanevsky, M. Kim
C.L. Liu and J.W. Layland (1973) provided a utilization bound test which is applicable to rate-monotonic priority assignment, when all tasks deadlines are at the end of their periods. Subsequently, J. Lehoczky et al. (1989) provided an exact schedulability criterion which is applicable to any arbitrary priority assignment scheme, with no restriction on task deadlines, which can be used when all task computation times are known exactly. In this work, we fill an important gap between these two, by presenting a technique for deriving utilization bounds, based on linear programming, that is applicable to similarly generalized situations, but does not require knowledge of exact task computation times.
{"title":"A generalized utilization bound test for fixed-priority real-time scheduling","authors":"Dong-Won Park, S. Natarajan, A. Kanevsky, M. Kim","doi":"10.1109/RTCSA.1995.528754","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528754","url":null,"abstract":"C.L. Liu and J.W. Layland (1973) provided a utilization bound test which is applicable to rate-monotonic priority assignment, when all tasks deadlines are at the end of their periods. Subsequently, J. Lehoczky et al. (1989) provided an exact schedulability criterion which is applicable to any arbitrary priority assignment scheme, with no restriction on task deadlines, which can be used when all task computation times are known exactly. In this work, we fill an important gap between these two, by presenting a technique for deriving utilization bounds, based on linear programming, that is applicable to similarly generalized situations, but does not require knowledge of exact task computation times.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"135 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133425513","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528775
Sung-June Hong
Among the many distinguishing features of the real-time computing system, the response time might be the most important one. In this paper we consider a MIMD type parallel computer system with different job priority and bursty job arrival pattern. Real-time jobs have higher priority than other jobs to guarantee faster response time, and the job server consists of array of processors, i.e. processor pool. Job requests arrive at different request buffers according to their priority, and we assume that the arrival pattern is a bursty process. These two issues, job priority and the bursty job request pattern, are studied for real-time computing systems. The processor pool is shared in such a way that if high priority queue is not empty, a fixed amount of the processing capacity is allocated to the queue, while the rest of the server is allocated to the other queue. Average waiting time is calculated for both queues. The optimal partition ratio of the processor pool is obtained analytically based on the average waiting time of high priority queue and low priority queue with respect to variable values of traffic loads /spl rho/.
{"title":"Optimal server allocation for real time computing systems with bursty priority jobs","authors":"Sung-June Hong","doi":"10.1109/RTCSA.1995.528775","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528775","url":null,"abstract":"Among the many distinguishing features of the real-time computing system, the response time might be the most important one. In this paper we consider a MIMD type parallel computer system with different job priority and bursty job arrival pattern. Real-time jobs have higher priority than other jobs to guarantee faster response time, and the job server consists of array of processors, i.e. processor pool. Job requests arrive at different request buffers according to their priority, and we assume that the arrival pattern is a bursty process. These two issues, job priority and the bursty job request pattern, are studied for real-time computing systems. The processor pool is shared in such a way that if high priority queue is not empty, a fixed amount of the processing capacity is allocated to the queue, while the rest of the server is allocated to the other queue. Average waiting time is calculated for both queues. The optimal partition ratio of the processor pool is obtained analytically based on the average waiting time of high priority queue and low priority queue with respect to variable values of traffic loads /spl rho/.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121224875","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528747
Junghoon Lee, Heonshik Shin
The objective of this paper is to develop a bandwidth reallocation scheme for TDMA-implemented Ethernet. The reallocation protocol consists of reporting change, calculating a new schedule, and reliably broadcasting the schedule. We also address the problem of static assignment of allocation vector, validate the essentiality of reallocation scheme, and design the scheme according to its functional requirements. The proposed scheme changes bandwidth allocation during the run-time operation of network system without affecting other traffics, so it can overcome the problem of static allocation. It completes reallocation within a bounded time as long as there are no transmission errors. Using this approach, the network is allowed to adapt to distributed real-time systems where traffic characteristics change dynamically.
{"title":"A bandwidth reallocation scheme for Ethernet-based real-time communication","authors":"Junghoon Lee, Heonshik Shin","doi":"10.1109/RTCSA.1995.528747","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528747","url":null,"abstract":"The objective of this paper is to develop a bandwidth reallocation scheme for TDMA-implemented Ethernet. The reallocation protocol consists of reporting change, calculating a new schedule, and reliably broadcasting the schedule. We also address the problem of static assignment of allocation vector, validate the essentiality of reallocation scheme, and design the scheme according to its functional requirements. The proposed scheme changes bandwidth allocation during the run-time operation of network system without affecting other traffics, so it can overcome the problem of static allocation. It completes reallocation within a bounded time as long as there are no transmission errors. Using this approach, the network is allowed to adapt to distributed real-time systems where traffic characteristics change dynamically.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115305320","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528761
Eun-Hee Hyun, Sung-Hee Kim
This paper presents practical and empirical approach to the main memory database system and proposes a minimal database system with real-time mobile data management functionality. It introduces a mobile computational viewpoint and gives a full detail of design considerations of real-time main memory database. As a case study, we present how to apply it to location registers in order to achieve both higher performance and a large volume of database capacity. It has a down-sized client/server architecture, including a relational data model, fast access methods, and middleware. This minimal real-time database system goes into use both mobility and subscriber management in a modern digital cellular system.
{"title":"Real-time mobile data management using a minimal MMDB","authors":"Eun-Hee Hyun, Sung-Hee Kim","doi":"10.1109/RTCSA.1995.528761","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528761","url":null,"abstract":"This paper presents practical and empirical approach to the main memory database system and proposes a minimal database system with real-time mobile data management functionality. It introduces a mobile computational viewpoint and gives a full detail of design considerations of real-time main memory database. As a case study, we present how to apply it to location registers in order to achieve both higher performance and a large volume of database capacity. It has a down-sized client/server architecture, including a relational data model, fast access methods, and middleware. This minimal real-time database system goes into use both mobility and subscriber management in a modern digital cellular system.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116641650","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528776
H. Fujita, T. Nakajima, H. Tezuka
Quality of service (QOS) is one of the most important factors in multimedia computing. Future continuous media applications should guarantee their QOS values provided by users, and support dynamic QOS control schemes by changing the QOS values according to system load. A processor reservation system developed in CMU is suitable for guaranteeing QOS, but it is not appropriate to support an application with a dynamic QOS control scheme. In this paper, we describe a new processor reservation system which is suitable for dynamic QOS control schemes. It consists of two components: user-level admission server and kernel support. The user-level admission server negotiates processor resources among applications, and the kernel support notifies the malicious use of processor resource by applications.
{"title":"A processor reservation system supporting dynamic QOS control","authors":"H. Fujita, T. Nakajima, H. Tezuka","doi":"10.1109/RTCSA.1995.528776","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528776","url":null,"abstract":"Quality of service (QOS) is one of the most important factors in multimedia computing. Future continuous media applications should guarantee their QOS values provided by users, and support dynamic QOS control schemes by changing the QOS values according to system load. A processor reservation system developed in CMU is suitable for guaranteeing QOS, but it is not appropriate to support an application with a dynamic QOS control scheme. In this paper, we describe a new processor reservation system which is suitable for dynamic QOS control schemes. It consists of two components: user-level admission server and kernel support. The user-level admission server negotiates processor resources among applications, and the kernel support notifies the malicious use of processor resource by applications.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130035250","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 : 1995-10-25DOI: 10.1109/RTCSA.1995.528764
K. Takashio, Hidehisa Shitomi, M. Tokoro
In this paper, the semantics of inter-object communication protocols supported by the distributed real-time programming language DROL that is an extension of C++ based on the Distributed Real-time Object (DRO) model is described. A characteristic common to these protocols is the introduction of a new notion, least suffering. The least suffering assures users of notification of network faults within a required timing constraint and rapid recovery from them. Thus, this notion allows users to construct real-time applications on widely distributed environments. It is impossible for previously proposed real-time programming languages to do it. Through the description of application programs and the discussion of their behavior, we show the effectiveness of the least suffering.
{"title":"Constructing distributed real-time systems with DROL real-time objects","authors":"K. Takashio, Hidehisa Shitomi, M. Tokoro","doi":"10.1109/RTCSA.1995.528764","DOIUrl":"https://doi.org/10.1109/RTCSA.1995.528764","url":null,"abstract":"In this paper, the semantics of inter-object communication protocols supported by the distributed real-time programming language DROL that is an extension of C++ based on the Distributed Real-time Object (DRO) model is described. A characteristic common to these protocols is the introduction of a new notion, least suffering. The least suffering assures users of notification of network faults within a required timing constraint and rapid recovery from them. Thus, this notion allows users to construct real-time applications on widely distributed environments. It is impossible for previously proposed real-time programming languages to do it. Through the description of application programs and the discussion of their behavior, we show the effectiveness of the least suffering.","PeriodicalId":111365,"journal":{"name":"Proceedings Second International Workshop on Real-Time Computing Systems and Applications","volume":"515 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1995-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123079804","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：