Pub Date : 1993-05-25DOI: 10.1109/ICDCS.1993.287704
Oliver E. Theel
Data replication schemes and mutual exclusion protocols can be regarded as special instances of cooperation schemes, which describe the interaction of independent nodes within a computer network to achieve a common goal. The author presents a new mechanism called general structured voting for cooperation management and demonstrates its use for handling instances of these problem domains. The proposed approach is shown to be very flexible, covers a wide range of scenarios, and supports an easy tailoring. In particular, it supports the switching from one cooperation scheme to another, e.g., as a reaction to changing network characteristics, simply by modifying the parameters of the model, avoiding a time and money consuming modification of the implementation.<>
{"title":"General structured voting: a flexible framework for modelling cooperations","authors":"Oliver E. Theel","doi":"10.1109/ICDCS.1993.287704","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287704","url":null,"abstract":"Data replication schemes and mutual exclusion protocols can be regarded as special instances of cooperation schemes, which describe the interaction of independent nodes within a computer network to achieve a common goal. The author presents a new mechanism called general structured voting for cooperation management and demonstrates its use for handling instances of these problem domains. The proposed approach is shown to be very flexible, covers a wide range of scenarios, and supports an easy tailoring. In particular, it supports the switching from one cooperation scheme to another, e.g., as a reaction to changing network characteristics, simply by modifying the parameters of the model, avoiding a time and money consuming modification of the implementation.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129940655","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287723
K. Shafer, Mohan L. Ahuja
High performance distributed computing systems require high performance communication systems. Distributed modeling and implementation of these communication systems is important. Toward this goal, the authors refine the process-to-channel/sub agent/-to-process (PCP) model of asynchronous distributed communication. While the PCP model provides a versatile and succinct mechanism for specifying and comparing different types of channels, it is inherently centralized. The refined model presented here, the process-to-channel/sub agent/-to-channel/sub agent/-to-process (PCCP) communication model, is amenable to distributed modeling and implementation of channels. The usefulness of the PCCP model is demonstrated by presenting a distributed implementation of hierarchical F-channels.<>
{"title":"Distributed modeling and implementation of high performance communication architectures","authors":"K. Shafer, Mohan L. Ahuja","doi":"10.1109/ICDCS.1993.287723","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287723","url":null,"abstract":"High performance distributed computing systems require high performance communication systems. Distributed modeling and implementation of these communication systems is important. Toward this goal, the authors refine the process-to-channel/sub agent/-to-process (PCP) model of asynchronous distributed communication. While the PCP model provides a versatile and succinct mechanism for specifying and comparing different types of channels, it is inherently centralized. The refined model presented here, the process-to-channel/sub agent/-to-channel/sub agent/-to-process (PCCP) communication model, is amenable to distributed modeling and implementation of channels. The usefulness of the PCCP model is demonstrated by presenting a distributed implementation of hierarchical F-channels.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132725010","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287724
H. Nishikawa, P. Steenkiste
The goal of load balancing is to assign to each node a number of tasks proportional to its performance. On distributed-memory machines, it is important to take data dependencies into account when distributing tasks, since they have a big impact on the communication requirements of the distributed application. The authors present a load balancing architecture that can deal with applications with heterogeneous tasks. The idea is to provide a set of load balancers that are effective for different types of homogeneous tasks, and to allow users to combine these load balancers for applications with heterogeneous tasks. This architecture was implemented on the Nectar multicomputer and performance results are presented for several applications with homogeneous and heterogeneous tasks.<>
{"title":"A general architecture for load balancing in a distributed-memory environment","authors":"H. Nishikawa, P. Steenkiste","doi":"10.1109/ICDCS.1993.287724","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287724","url":null,"abstract":"The goal of load balancing is to assign to each node a number of tasks proportional to its performance. On distributed-memory machines, it is important to take data dependencies into account when distributing tasks, since they have a big impact on the communication requirements of the distributed application. The authors present a load balancing architecture that can deal with applications with heterogeneous tasks. The idea is to provide a set of load balancers that are effective for different types of homogeneous tasks, and to allow users to combine these load balancers for applications with heterogeneous tasks. This architecture was implemented on the Nectar multicomputer and performance results are presented for several applications with homogeneous and heterogeneous tasks.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126579198","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287693
D. Long, Carol Osterbrock, L. Cabrera
A network subsystem supporting a continuous media file system must guarantee a minimum throughput, a maximum delay, and a maximum jitter. The authors present a transport protocol that provides these guarantees. To support different types of service, the protocol is built from modules selected to meet the requirements of each communication session. A buffering technique is used to provide jitter guarantees. To provide throughput and delay guarantees, network performance is optimized based on the required transfer rate. The effects of controlling transmission rate and packet size are presented. The resulting transport protocol is modeled on a simulated FDDI (fiber distributed data interface) network and the results are analyzed. It is shown that the protocol provides the required guarantees for the anticipated types of traffic.<>
{"title":"Providing performance guarantees in an FDDI network","authors":"D. Long, Carol Osterbrock, L. Cabrera","doi":"10.1109/ICDCS.1993.287693","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287693","url":null,"abstract":"A network subsystem supporting a continuous media file system must guarantee a minimum throughput, a maximum delay, and a maximum jitter. The authors present a transport protocol that provides these guarantees. To support different types of service, the protocol is built from modules selected to meet the requirements of each communication session. A buffering technique is used to provide jitter guarantees. To provide throughput and delay guarantees, network performance is optimized based on the required transfer rate. The effects of controlling transmission rate and packet size are presented. The resulting transport protocol is modeled on a simulated FDDI (fiber distributed data interface) network and the results are analyzed. It is shown that the protocol provides the required guarantees for the anticipated types of traffic.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"156 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131937198","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287690
Ming-Syan Chen, Philip S. Yu, Kun-Lung Wu
The authors develop efficient decentralized consensus protocols for a distributed system with multi-port communication. Two classes of decentralized consensus protocols are considered: the one without an initiator and the one with an initiator. The case of one-port communication is first presented, i.e., each node can send out one message in one step, and then results are derived for the case of multi-port communication, i.e., each node can send out more than one message in one step. Given an arbitrary number of nodes in a system, the proposed protocols can reach the consensus in the minimal numbers of message steps. The number of messages incurred by each algorithm is also derived.<>
{"title":"Decentralized consensus protocols with multi-port communication","authors":"Ming-Syan Chen, Philip S. Yu, Kun-Lung Wu","doi":"10.1109/ICDCS.1993.287690","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287690","url":null,"abstract":"The authors develop efficient decentralized consensus protocols for a distributed system with multi-port communication. Two classes of decentralized consensus protocols are considered: the one without an initiator and the one with an initiator. The case of one-port communication is first presented, i.e., each node can send out one message in one step, and then results are derived for the case of multi-port communication, i.e., each node can send out more than one message in one step. Given an arbitrary number of nodes in a system, the proposed protocols can reach the consensus in the minimal numbers of message steps. The number of messages incurred by each algorithm is also derived.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132459530","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287729
Darryl L. Willick, D. Eager, R. Bunt
Trace driven simulations were used to study the performance of several disk cache replacement policies for network file servers. It is shown that locality based approaches, such as the common least recently used (LRU) policy, which are known to work well on stand-alone disked workstations and at client workstations in distributed systems, are inappropriate at a fileserver. Quite simple frequency based approaches do better. More sophisticated frequency based policies (eg., that take into account the file type) may offer additional improvements.<>
{"title":"Disk cache replacement policies for network fileservers","authors":"Darryl L. Willick, D. Eager, R. Bunt","doi":"10.1109/ICDCS.1993.287729","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287729","url":null,"abstract":"Trace driven simulations were used to study the performance of several disk cache replacement policies for network file servers. It is shown that locality based approaches, such as the common least recently used (LRU) policy, which are known to work well on stand-alone disked workstations and at client workstations in distributed systems, are inappropriate at a fileserver. Quite simple frequency based approaches do better. More sophisticated frequency based policies (eg., that take into account the file type) may offer additional improvements.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124899368","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287699
H. Gladney
Emerging technology permits electronic document libraries which store large numbers of images for access in wide-area networks. The expected statistics and security needs of electronic library applications favor new client-server session protocols. The author proposes a scheme that makes communications efficient in a well-defined sense. DocSS, a document storage subsystem, implements the scheme. An early version is in production in human service (welfare, taxation, police) and engineering applications.<>
{"title":"Inter-machine protocols for electronic libraries","authors":"H. Gladney","doi":"10.1109/ICDCS.1993.287699","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287699","url":null,"abstract":"Emerging technology permits electronic document libraries which store large numbers of images for access in wide-area networks. The expected statistics and security needs of electronic library applications favor new client-server session protocols. The author proposes a scheme that makes communications efficient in a well-defined sense. DocSS, a document storage subsystem, implements the scheme. An early version is in production in human service (welfare, taxation, police) and engineering applications.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123555995","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287672
Chang-Jia Wang, P. Krueger, Ming T. Liu
A key issue in distributed scheduling is selecting appropriate jobs to transfer. A job selection policy that considers the diversity of job behaviors is proposed. A mechanism used in artificial neural networks, called weight climbing, is employed. Using this mechanism, a distributed scheduler can learn the behavior of a job from its past executions and make a correct prediction about whether transferring the job is worthwhile. A scheduler using the proposed job selection policy has been implemented and experimental results show that it is able to learn job behaviors fast, make decisions accurately and adjust itself promptly when system configuration or program behaviors are changed. In addition, the selection policy introduces only negligible time and space overhead.<>
{"title":"Intelligent job selection for distributed scheduling","authors":"Chang-Jia Wang, P. Krueger, Ming T. Liu","doi":"10.1109/ICDCS.1993.287672","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287672","url":null,"abstract":"A key issue in distributed scheduling is selecting appropriate jobs to transfer. A job selection policy that considers the diversity of job behaviors is proposed. A mechanism used in artificial neural networks, called weight climbing, is employed. Using this mechanism, a distributed scheduler can learn the behavior of a job from its past executions and make a correct prediction about whether transferring the job is worthwhile. A scheduler using the proposed job selection policy has been implemented and experimental results show that it is able to learn job behaviors fast, make decisions accurately and adjust itself promptly when system configuration or program behaviors are changed. In addition, the selection policy introduces only negligible time and space overhead.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129680548","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287673
B. Millard, P. Dasgupta, Sanjay G. Rao, R. Kuramkote
The authors present the design and implementation of a persistent store called SPOMS. SPOMS is a runtime system that provides a store for persistent objects and is language independent. The objects are created via calls to SPOMS, and, when used, SPOMS directly maps them into the spaces of all requesting processes. The objects are stored in native format and are concurrently sharable. The store can handle distributed applications. The system uses the concept of a compiled class to manage persistent objects. The compiled class is a template that is used to create and store objects in a language independent manner and so that object reuse can occur without recompilation or relinking of an application that uses it. A prototype of SPOMS has been built on top of the Mach operating system. The motivations, the design, and implementation details are presented. Related and future work are discussed.<>
{"title":"Run-time support and storage management for memory-mapped persistent objects","authors":"B. Millard, P. Dasgupta, Sanjay G. Rao, R. Kuramkote","doi":"10.1109/ICDCS.1993.287673","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287673","url":null,"abstract":"The authors present the design and implementation of a persistent store called SPOMS. SPOMS is a runtime system that provides a store for persistent objects and is language independent. The objects are created via calls to SPOMS, and, when used, SPOMS directly maps them into the spaces of all requesting processes. The objects are stored in native format and are concurrently sharable. The store can handle distributed applications. The system uses the concept of a compiled class to manage persistent objects. The compiled class is a template that is used to create and store objects in a language independent manner and so that object reuse can occur without recompilation or relinking of an application that uses it. A prototype of SPOMS has been built on top of the Mach operating system. The motivations, the design, and implementation details are presented. Related and future work are discussed.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116698742","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 : 1993-05-25DOI: 10.1109/ICDCS.1993.287712
A. Ghedamsi, G. Bochmann, R. Dssouli
The authors propose a generalized diagnostic algorithm for the case where more than one fault (output or transfer) may be present in one of the transitions of a deterministic system represented by a set of communicating finite state machines (CFSMs). Such an algorithm localizes the faulty transition in the distributed system once the fault has been detected. It generates, if necessary, additional diagnostic test cases which depend on the observed symptoms and which permit the location of the detected faults. The algorithm guarantees the correct diagnosis of any single or double fault (output and/or transfer) in at most one of the transitions of a deterministic system which is represented by a set of communicating FSMs. A simple example is used to demonstrate the functioning of the different steps of the proposed diagnostic algorithm.<>
{"title":"Diagnosis of single transition faults in communicating finite state machines","authors":"A. Ghedamsi, G. Bochmann, R. Dssouli","doi":"10.1109/ICDCS.1993.287712","DOIUrl":"https://doi.org/10.1109/ICDCS.1993.287712","url":null,"abstract":"The authors propose a generalized diagnostic algorithm for the case where more than one fault (output or transfer) may be present in one of the transitions of a deterministic system represented by a set of communicating finite state machines (CFSMs). Such an algorithm localizes the faulty transition in the distributed system once the fault has been detected. It generates, if necessary, additional diagnostic test cases which depend on the observed symptoms and which permit the location of the detected faults. The algorithm guarantees the correct diagnosis of any single or double fault (output and/or transfer) in at most one of the transitions of a deterministic system which is represented by a set of communicating FSMs. A simple example is used to demonstrate the functioning of the different steps of the proposed diagnostic algorithm.<<ETX>>","PeriodicalId":249060,"journal":{"name":"[1993] Proceedings. The 13th International Conference on Distributed Computing Systems","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1993-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117170868","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：