Pub Date : 1996-05-27DOI: 10.1109/ICDCS.1996.507910
Chi-Che Chou, K. Shin
Given the client's traffic-generation characteristics and performance requirements, we propose a real-time communication scheme that provides delivery delay guarantees in a multiaccess local-area network (LAN). This scheme (i) reduces the link capacity that needs to be reserved to an average level as compared to the worst-case level required for deterministic performance guarantees, and (ii) preserves the ability of independent addition and deletion of real-time channels.
{"title":"Multiplexing statistical real-time channels in a multi-access network","authors":"Chi-Che Chou, K. Shin","doi":"10.1109/ICDCS.1996.507910","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507910","url":null,"abstract":"Given the client's traffic-generation characteristics and performance requirements, we propose a real-time communication scheme that provides delivery delay guarantees in a multiaccess local-area network (LAN). This scheme (i) reduces the link capacity that needs to be reserved to an average level as compared to the worst-case level required for deterministic performance guarantees, and (ii) preserves the ability of independent addition and deletion of real-time channels.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133882749","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507918
B. Hamidzadeh, D. Lilja
Efficiently scheduling parallel tasks on to the processors of a shared-memory multiprocessor is critical to achieving high performance. Given perfect information at compile-time, a static scheduling strategy can produce an assignment of tasks to processors that ideally balances the load among the processors while minimizing the run-time scheduling overhead and the average memory referencing delay. Since perfect information is seldom available, however, dynamic scheduling strategies distribute the task assignment function to the processors by having idle processors allocate work to themselves from a shared queue. While this approach can improve the load balancing compared to static scheduling, the time required to access the shared work queue adds directly to the overall execution time. To overlap the time required to dynamically schedule tasks with the execution of the tasks, we examine a class of self-adjusting dynamic scheduling (SADS) algorithms that centralizes the assignment of tasks to processors. These algorithms dedicate a single processor of the multiprocessor to perform a novel on-line branch-and-bound technique that dynamically computes partial schedules based on the loads of the other processors and the memory locality (affinity) of the tasks and the processors. Our simulation results show that this centralized scheduling outperforms self-scheduling algorithms even when using only a small number of processors.
{"title":"Dynamic scheduling strategies for shared-memory multiprocessors","authors":"B. Hamidzadeh, D. Lilja","doi":"10.1109/ICDCS.1996.507918","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507918","url":null,"abstract":"Efficiently scheduling parallel tasks on to the processors of a shared-memory multiprocessor is critical to achieving high performance. Given perfect information at compile-time, a static scheduling strategy can produce an assignment of tasks to processors that ideally balances the load among the processors while minimizing the run-time scheduling overhead and the average memory referencing delay. Since perfect information is seldom available, however, dynamic scheduling strategies distribute the task assignment function to the processors by having idle processors allocate work to themselves from a shared queue. While this approach can improve the load balancing compared to static scheduling, the time required to access the shared work queue adds directly to the overall execution time. To overlap the time required to dynamically schedule tasks with the execution of the tasks, we examine a class of self-adjusting dynamic scheduling (SADS) algorithms that centralizes the assignment of tasks to processors. These algorithms dedicate a single processor of the multiprocessor to perform a novel on-line branch-and-bound technique that dynamically computes partial schedules based on the loads of the other processors and the memory locality (affinity) of the tasks and the processors. Our simulation results show that this centralized scheduling outperforms self-scheduling algorithms even when using only a small number of processors.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133214898","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507926
D. Hagimont, J. Mossière, X. R. D. Pina, F. Saunier
Software capabilities are a very convenient means to protect co-operating applications. They allow access rights to be dynamically exchanged between mutually suspicious interacting applications. However in all the proposed approaches, capabilities are made available at the programming language level, requiring application developers to wire protection definition in the application code, which is detrimental to both flexibility and reusability. We believe instead that capabilities should be hidden from the application programmer allowing protection definition and application code to be clearly separated. In this paper we propose a new protection model based on hidden software capabilities, in which protection definition is completely disjoined from the application code and described in an extended interface definition language (IDL). This allows to specify protection for existing modules and to easily change the protection policy of an application. This protection model can be integrated in a wide range of operating systems. We are currently implementing it in a single address space operating system based on distributed shared memory.
{"title":"Hidden software capabilities","authors":"D. Hagimont, J. Mossière, X. R. D. Pina, F. Saunier","doi":"10.1109/ICDCS.1996.507926","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507926","url":null,"abstract":"Software capabilities are a very convenient means to protect co-operating applications. They allow access rights to be dynamically exchanged between mutually suspicious interacting applications. However in all the proposed approaches, capabilities are made available at the programming language level, requiring application developers to wire protection definition in the application code, which is detrimental to both flexibility and reusability. We believe instead that capabilities should be hidden from the application programmer allowing protection definition and application code to be clearly separated. In this paper we propose a new protection model based on hidden software capabilities, in which protection definition is completely disjoined from the application code and described in an extended interface definition language (IDL). This allows to specify protection for existing modules and to easily change the protection policy of an application. This protection model can be integrated in a wide range of operating systems. We are currently implementing it in a single address space operating system based on distributed shared memory.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125178835","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.508008
L. Bellissard, S. Atallah, F. Boyer, M. Riveill
This paper presents Olan, a language and a run time support intended to facilitate the design, configuration and evolution of distributed applications made up of heterogeneous software components. Configuration covers two phases: for the application builder the identification of the software components and the description of their interconnections and communications; for the application administrator and maintainer the accurate use of system resources provided by the target environment, such as the placement of components on nodes. The main benefit of Olan is to provide a single unified description of distributed applications, adequate for construction, management, and evolution. The overall description is independent from the components' implementation, so that the configuration process, e.g. the production of different versions of an implementation, is decoupled from the programming process. The features of the configuration language and the functions of the runtime support system are illustrated through the example of a distributed teleconferencing application.
{"title":"Distributed application configuration","authors":"L. Bellissard, S. Atallah, F. Boyer, M. Riveill","doi":"10.1109/ICDCS.1996.508008","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.508008","url":null,"abstract":"This paper presents Olan, a language and a run time support intended to facilitate the design, configuration and evolution of distributed applications made up of heterogeneous software components. Configuration covers two phases: for the application builder the identification of the software components and the description of their interconnections and communications; for the application administrator and maintainer the accurate use of system resources provided by the target environment, such as the placement of components on nodes. The main benefit of Olan is to provide a single unified description of distributed applications, adequate for construction, management, and evolution. The overall description is independent from the components' implementation, so that the configuration process, e.g. the production of different versions of an implementation, is decoupled from the programming process. The features of the configuration language and the functions of the runtime support system are illustrated through the example of a distributed teleconferencing application.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121940540","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.508014
Chin Lu, S. Lau
We propose an adaptive load balancing algorithm for heterogeneous distributed systems. The algorithm intrinsically allows a batch of tasks to be relocated. The key of the algorithm is to transfer a suitable amount of processing demand from senders to receivers. This amount is determined dynamically during sender-receiver negotiations. Factors considered when this amount is determined include processing speeds of different nodes, the current load state of both sender and receiver, and the processing demands of tasks eligible for relocation. Composition of a task batch is modeled as a 0-1 Knapsack problem. We also propose a load state measurement scheme which is designed particularly for heterogeneous systems.
{"title":"An adaptive load balancing algorithm for heterogeneous distributed systems with multiple task classes","authors":"Chin Lu, S. Lau","doi":"10.1109/ICDCS.1996.508014","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.508014","url":null,"abstract":"We propose an adaptive load balancing algorithm for heterogeneous distributed systems. The algorithm intrinsically allows a batch of tasks to be relocated. The key of the algorithm is to transfer a suitable amount of processing demand from senders to receivers. This amount is determined dynamically during sender-receiver negotiations. Factors considered when this amount is determined include processing speeds of different nodes, the current load state of both sender and receiver, and the processing demands of tasks eligible for relocation. Composition of a task batch is modeled as a 0-1 Knapsack problem. We also propose a load state measurement scheme which is designed particularly for heterogeneous systems.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122374165","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.508006
H. Nakajo, S. Ohtani, Y. Kaneda
JUMP-1 is a distributed shared-memory massively parallel computer and is composed of multiple clusters of inter-connected network called RDT (Recursive Diagonal Torus). Each cluster in JUMP-1 consists of 4 element processors, secondary cache memories, and 2 MBP (Memory Based Processor) for high-speed synchronization and communication among clusters. The I/O subsystem is connected to a cluster via a high-speed serial link called STAFF-Link. The I/O buffer memory is mapped onto the JUMP-1 global shared-memory to permit each I/O access operation as memory access. In this paper we describe evaluation of the fundamental performance of the disk I/O subsystem using event-driven simulation, and estimated performance with a Video On Demand (VOD) application.
{"title":"A simulation-based evaluation of a disk I/O subsystem for a massively parallel computer: JUMP-1","authors":"H. Nakajo, S. Ohtani, Y. Kaneda","doi":"10.1109/ICDCS.1996.508006","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.508006","url":null,"abstract":"JUMP-1 is a distributed shared-memory massively parallel computer and is composed of multiple clusters of inter-connected network called RDT (Recursive Diagonal Torus). Each cluster in JUMP-1 consists of 4 element processors, secondary cache memories, and 2 MBP (Memory Based Processor) for high-speed synchronization and communication among clusters. The I/O subsystem is connected to a cluster via a high-speed serial link called STAFF-Link. The I/O buffer memory is mapped onto the JUMP-1 global shared-memory to permit each I/O access operation as memory access. In this paper we describe evaluation of the fundamental performance of the disk I/O subsystem using event-driven simulation, and estimated performance with a Video On Demand (VOD) application.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133730314","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507903
Shih-Chen Huang, Z. Kedem
We introduce a shared memory software prototype system for executing programs with nested parallelism on a network of workstations. This programming model exhibits a very convenient and natural programming style and provides functionality similar to a subset of Compositional C++. Such programming model is especially suitable for computations whose complexity and parallelism emerges only during their execution, as in divide and conquer problems. To both support and take advantage of the flexibility inherent in the programming model, we develop an architecture, which distributes both the shared memory management and the computation, removing bottlenecks inherent in centralization, thus also providing scalability and dependability. The system supports also dynamic load balancing, and fault tolerance-both transparently to the programmer. The prototype performs well using the realistic platforms of non-dedicated network of workstation. We describe encouraging performance experiments on a network in which some of the machines became slow unpredictably (to the application program). The system coped well with such dynamic behavior.
{"title":"Supporting a flexible parallel programming model on a network of workstations","authors":"Shih-Chen Huang, Z. Kedem","doi":"10.1109/ICDCS.1996.507903","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507903","url":null,"abstract":"We introduce a shared memory software prototype system for executing programs with nested parallelism on a network of workstations. This programming model exhibits a very convenient and natural programming style and provides functionality similar to a subset of Compositional C++. Such programming model is especially suitable for computations whose complexity and parallelism emerges only during their execution, as in divide and conquer problems. To both support and take advantage of the flexibility inherent in the programming model, we develop an architecture, which distributes both the shared memory management and the computation, removing bottlenecks inherent in centralization, thus also providing scalability and dependability. The system supports also dynamic load balancing, and fault tolerance-both transparently to the programmer. The prototype performs well using the realistic platforms of non-dedicated network of workstation. We describe encouraging performance experiments on a network in which some of the machines became slow unpredictably (to the application program). The system coped well with such dynamic behavior.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128237134","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507989
M. Silva, M. Atkinson, A. Black
Current RPC mechanisms for persistent languages are either pass by reference-in which case they do not scale-or pass by copy-in which case they duplicate objects and destroy sharing relationships. In this paper we argue that to build very large distributed persistent applications a compromise between these two mechanisms is needed. The ultimate goal of our research is to build a scalable persistent RPC while still maintaining object sharing type safety, type completeness and semantics that are readily understood by application programmers.
{"title":"Semantics for parameter passing in a type-complete persistent RPC","authors":"M. Silva, M. Atkinson, A. Black","doi":"10.1109/ICDCS.1996.507989","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507989","url":null,"abstract":"Current RPC mechanisms for persistent languages are either pass by reference-in which case they do not scale-or pass by copy-in which case they duplicate objects and destroy sharing relationships. In this paper we argue that to build very large distributed persistent applications a compromise between these two mechanisms is needed. The ultimate goal of our research is to build a scalable persistent RPC while still maintaining object sharing type safety, type completeness and semantics that are readily understood by application programmers.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117330952","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507985
K. Karlapalem, Qing Li, S. Vieweg
Object-oriented database systems are becoming popular and are being used in a large number of application domain. Many of these application domains are inherently distributed. The focus of this work is on articulating the concepts of method induced partitioning schemes in object-oriented databases by understanding and classifying the object behavior embodied by the methods. We provide a solution for supporting fragmentation transparency by using method transformation. Finally, we present guidelines for method induced partitioning in object-oriented databases.
{"title":"Method-induced partitioning schemes for object-oriented databases","authors":"K. Karlapalem, Qing Li, S. Vieweg","doi":"10.1109/ICDCS.1996.507985","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507985","url":null,"abstract":"Object-oriented database systems are becoming popular and are being used in a large number of application domain. Many of these application domains are inherently distributed. The focus of this work is on articulating the concepts of method induced partitioning schemes in object-oriented databases by understanding and classifying the object behavior embodied by the methods. We provide a solution for supporting fragmentation transparency by using method transformation. Finally, we present guidelines for method induced partitioning in object-oriented databases.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133494054","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 : 1996-05-27DOI: 10.1109/ICDCS.1996.507993
A. Tomasic, L. Raschid, P. Valduriez
Access to large numbers of data sources introduces new problems for users of heterogeneous distributed databases. End users and application programmers must deal with unavailable data sources. Database administrators must deal with incorporating new sources into the model. Database implementers must deal with the translation of queries between query languages and schemas. The Distributed Information Search COmponent (Disco) addresses these problems. Query processing semantics are developed to process queries over data sources which do not return answers. Data modeling techniques manage connections to data sources. The component interface to data sources flexibly handles different query languages and translates queries. This paper describes (a) the distributed mediator architecture of Disco, (b) its query processing semantics, (C) the data model and its modeling of data source connections, and (d) the interface to underlying data sources.
{"title":"Scaling heterogeneous databases and the design of Disco","authors":"A. Tomasic, L. Raschid, P. Valduriez","doi":"10.1109/ICDCS.1996.507993","DOIUrl":"https://doi.org/10.1109/ICDCS.1996.507993","url":null,"abstract":"Access to large numbers of data sources introduces new problems for users of heterogeneous distributed databases. End users and application programmers must deal with unavailable data sources. Database administrators must deal with incorporating new sources into the model. Database implementers must deal with the translation of queries between query languages and schemas. The Distributed Information Search COmponent (Disco) addresses these problems. Query processing semantics are developed to process queries over data sources which do not return answers. Data modeling techniques manage connections to data sources. The component interface to data sources flexibly handles different query languages and translates queries. This paper describes (a) the distributed mediator architecture of Disco, (b) its query processing semantics, (C) the data model and its modeling of data source connections, and (d) the interface to underlying data sources.","PeriodicalId":159322,"journal":{"name":"Proceedings of 16th International Conference on Distributed Computing Systems","volume":"126 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1996-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120858347","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：