There is increasing demand to extend object request broker (ORB) middleware to support applications with stringent quality of service (QoS) requirements. However conventional ORBs do not define standard features for specifying or enforcing end-to-end QoS for applications with deterministic real-time requirements. This paper describes the design and performance of a real-time I/O (RIO) subsystem optimized for QoS-enabled ORB end-systems that support high-performance and real-time applications running on off-the-shelf hardware and software. The paper illustrates how integrating a real-time ORB with a RIO subsystem can reduce latency bounds on end-to-end communication between high-priority clients without unduly penalizing low-priority and best-effort clients.
{"title":"The performance of a real-time I/O subsystem for QoS-enabled ORB middleware","authors":"F. Kuhns, D. Schmidt, D. Levine","doi":"10.1109/DOA.1999.793998","DOIUrl":"https://doi.org/10.1109/DOA.1999.793998","url":null,"abstract":"There is increasing demand to extend object request broker (ORB) middleware to support applications with stringent quality of service (QoS) requirements. However conventional ORBs do not define standard features for specifying or enforcing end-to-end QoS for applications with deterministic real-time requirements. This paper describes the design and performance of a real-time I/O (RIO) subsystem optimized for QoS-enabled ORB end-systems that support high-performance and real-time applications running on off-the-shelf hardware and software. The paper illustrates how integrating a real-time ORB with a RIO subsystem can reduce latency bounds on end-to-end communication between high-priority clients without unduly penalizing low-priority and best-effort clients.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115946156","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}
Transaction management is required to maintain data consistency in any system with concurrent access to its information. The Object Transaction Service (OTS) is a CORBA service that brings transactional semantics to distributed object oriented applications. We report our experience of developing a prototype using OTS. We present results from our scaling experiments and analyze the overheads incurred by the components of an OTS based distributed transaction. We also highlight issues that need to be addressed if OTS is to be used in large distributed systems such as those in Boeing.
{"title":"Object Transaction Service: experiences and open issues","authors":"P. Ram, L. Do, P. Drew, Tong Zhou","doi":"10.1109/DOA.1999.794049","DOIUrl":"https://doi.org/10.1109/DOA.1999.794049","url":null,"abstract":"Transaction management is required to maintain data consistency in any system with concurrent access to its information. The Object Transaction Service (OTS) is a CORBA service that brings transactional semantics to distributed object oriented applications. We report our experience of developing a prototype using OTS. We present results from our scaling experiments and analyze the overheads incurred by the components of an OTS based distributed transaction. We also highlight issues that need to be addressed if OTS is to be used in large distributed systems such as those in Boeing.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129016756","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}
T. Plagemann, F. Eliassen, V. Goebel, Tom Kristensen, Hans Ole Rafaelsen
The heterogeneity of distributed multimedia systems in terms of hardware, operating systems, programming languages, data formats, compression formats, available resources, and QoS requirements imposes severe problems. The purpose of the middleware platform is to provide interoperability and portability of distributed system services and applications. However, today's middleware platforms, like CORBA and TINA-DPE, do not solve all these problems. For example, automated control of interface compatibility and quality of service (QoS) support are still important research topics. This is especially true in the context of multimedia database systems (MMDBS) in distributed systems, because interface type and QoS depends on the result of queries and cannot be determined in advance. We show how the temporal object oriented data model, TOOMM, and the formal model for bindings MBS can be utilized in a heterogeneous distributed system to: (1) solve compatibility issues, (2) find agreeable levels of QoS, and (3) automatically adapt the flexible multimedia object request broker (ORB) MULTE-ORB, to application requirements and available resources.
分布式多媒体系统在硬件、操作系统、编程语言、数据格式、压缩格式、可用资源和QoS要求方面的异构性带来了严重的问题。中间件平台的目的是提供分布式系统服务和应用程序的互操作性和可移植性。然而，今天的中间件平台，如CORBA和TINA-DPE，并不能解决所有这些问题。例如，接口兼容性的自动控制和服务质量(QoS)支持仍然是重要的研究课题。在分布式系统中的多媒体数据库系统(MMDBS)环境中尤其如此，因为接口类型和QoS依赖于查询的结果，不能提前确定。我们将展示如何在异构分布式系统中利用面向对象的时态数据模型tomm和绑定MBS的正式模型:(1)解决兼容性问题，(2)找到合适的QoS级别，以及(3)自动调整灵活的多媒体对象请求代理(ORB) multi -ORB，以适应应用程序需求和可用资源。
{"title":"Adaptive QoS aware binding of persistent multimedia objects","authors":"T. Plagemann, F. Eliassen, V. Goebel, Tom Kristensen, Hans Ole Rafaelsen","doi":"10.1109/DOA.1999.794052","DOIUrl":"https://doi.org/10.1109/DOA.1999.794052","url":null,"abstract":"The heterogeneity of distributed multimedia systems in terms of hardware, operating systems, programming languages, data formats, compression formats, available resources, and QoS requirements imposes severe problems. The purpose of the middleware platform is to provide interoperability and portability of distributed system services and applications. However, today's middleware platforms, like CORBA and TINA-DPE, do not solve all these problems. For example, automated control of interface compatibility and quality of service (QoS) support are still important research topics. This is especially true in the context of multimedia database systems (MMDBS) in distributed systems, because interface type and QoS depends on the result of queries and cannot be determined in advance. We show how the temporal object oriented data model, TOOMM, and the formal model for bindings MBS can be utilized in a heterogeneous distributed system to: (1) solve compatibility issues, (2) find agreeable levels of QoS, and (3) automatically adapt the flexible multimedia object request broker (ORB) MULTE-ORB, to application requirements and available resources.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128312851","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}
R. Durrant, G. J. Sweeney, M. Fleury, A. Downton, A. Clark
Pipelined Processor Farms (PPF) are a generic parallel processing structure suitable for continuous flow, multi algorithm applications. PPF has been redesigned for a multimedia workload which is multifunctional and multimodal. A dynamic object based processing structure is implemented which will map onto mobile stations. The PPF processing model with asynchronous message passing in a multi threaded environment is in effect a version of the actor paradigm which is a dynamic form of encapsulated system. Stable software facilities for distributed object based systems which may include reflection are also needed. The supportive facilities of Java remote method invocation (RMI) are currently being used by PPF. PPF and RMI have been wedded together in a software toolkit, providing a development environment which includes performance prediction, analysis, and guided code generation.
{"title":"Providing structure for medium-grained distributed object-based computation","authors":"R. Durrant, G. J. Sweeney, M. Fleury, A. Downton, A. Clark","doi":"10.1109/DOA.1999.794045","DOIUrl":"https://doi.org/10.1109/DOA.1999.794045","url":null,"abstract":"Pipelined Processor Farms (PPF) are a generic parallel processing structure suitable for continuous flow, multi algorithm applications. PPF has been redesigned for a multimedia workload which is multifunctional and multimodal. A dynamic object based processing structure is implemented which will map onto mobile stations. The PPF processing model with asynchronous message passing in a multi threaded environment is in effect a version of the actor paradigm which is a dynamic form of encapsulated system. Stable software facilities for distributed object based systems which may include reflection are also needed. The supportive facilities of Java remote method invocation (RMI) are currently being used by PPF. PPF and RMI have been wedded together in a software toolkit, providing a development environment which includes performance prediction, analysis, and guided code generation.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122354580","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}
L. Silva, Victor Batista, Paulo Martins, Guilherme Soares
Mobile code is a promising model for distributed computing and it has been exploited in several areas of applications. One of the areas that may benefit from the use of mobile agent technology is parallel processing. This paper describes a Java-based platform, called JAMES, that provides support for parallel computing. We have implemented a software module that supports the well-known model of master/worker and we have exploited the use of parallel computing in some distributed tasks. We present some experimental results that compare the master/worker model with the usual model of migratory agents. Then, we compare the use of mobile agents with two other solutions for parallel computing: MPI and JET. The results are quite promising.
{"title":"Using mobile agents for parallel processing","authors":"L. Silva, Victor Batista, Paulo Martins, Guilherme Soares","doi":"10.1109/DOA.1999.793980","DOIUrl":"https://doi.org/10.1109/DOA.1999.793980","url":null,"abstract":"Mobile code is a promising model for distributed computing and it has been exploited in several areas of applications. One of the areas that may benefit from the use of mobile agent technology is parallel processing. This paper describes a Java-based platform, called JAMES, that provides support for parallel computing. We have implemented a software module that supports the well-known model of master/worker and we have exploited the use of parallel computing in some distributed tasks. We present some experimental results that compare the master/worker model with the usual model of migratory agents. Then, we compare the use of mobile agents with two other solutions for parallel computing: MPI and JET. The results are quite promising.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116897801","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}
Discusses how to replicate a server in order to obtain fault-tolerant services. In active replication, all the requests from a client are performed by all the server replicas in the same order. The replicas are placed on heterogeneous computers interconnected in a wide area network, e.g. the Internet, where each communication channel has a different message transmission delay and bandwidth. Hence, the response times of the replicas observed by the clients are not the same. We propose a pseudo-active replication, where the client does not wait for other responses after receiving one response, and slower replicas do not perform every request performed by the faster replicas.
{"title":"Pseudo-active replication of distributed objects in wide-area networks","authors":"H. Higaki, Katsuya Tanaka, M. Takizawa","doi":"10.1109/DOA.1999.794004","DOIUrl":"https://doi.org/10.1109/DOA.1999.794004","url":null,"abstract":"Discusses how to replicate a server in order to obtain fault-tolerant services. In active replication, all the requests from a client are performed by all the server replicas in the same order. The replicas are placed on heterogeneous computers interconnected in a wide area network, e.g. the Internet, where each communication channel has a different message transmission delay and bandwidth. Hence, the response times of the replicas observed by the clients are not the same. We propose a pseudo-active replication, where the client does not wait for other responses after receiving one response, and slower replicas do not perform every request performed by the faster replicas.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116961942","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}
Existing traders support the required functions, such as the lookup and registry of services. They are limited because they are restricted to a small number of services. The paper proposes solutions to the issues of persistence and query routing in the context of CORBA Trader to improve the performance of the core trader functions. We propose appropriate wrappers that provide database-like interfaces for the traders. This enables transparency and efficient access to persistent services. We propose a design of a full CORBA trader implementation, called DOK-Trader. It has specific functionalities, since it includes a specific routing mechanism for the trader federated queries. We demonstrate that these functionalities have led to better performance for the core CORBA trader functions.
{"title":"DOK-Trader: a CORBA persistent trader with query routing facilities","authors":"G. Craske, Z. Tari, K. Kumar","doi":"10.1109/DOA.1999.794035","DOIUrl":"https://doi.org/10.1109/DOA.1999.794035","url":null,"abstract":"Existing traders support the required functions, such as the lookup and registry of services. They are limited because they are restricted to a small number of services. The paper proposes solutions to the issues of persistence and query routing in the context of CORBA Trader to improve the performance of the core trader functions. We propose appropriate wrappers that provide database-like interfaces for the traders. This enables transparency and efficient access to persistent services. We propose a design of a full CORBA trader implementation, called DOK-Trader. It has specific functionalities, since it includes a specific routing mechanism for the trader federated queries. We demonstrate that these functionalities have led to better performance for the core CORBA trader functions.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125790211","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}
Active systems have been proposed as a paradigm to satisfy the needs of many databases and other applications that require a timely response to situations. Event-Condition-Action (or ECA) rules (S. Chakravarthy et al., 1989) are used to capture the active capability in a system. The utility and functionality of active capability (ECA roles) has been well established in the context of databases. In order for the active capability to be useful for a large class of advanced applications, it is necessary to go beyond what has been proposed/developed in the context of databases. Specifically, extensions beyond the current state of the art in active capability are needed along several dimensions: i) make the active capability available for non-database applications, in addition to database applications; ii) make the active capability available in distributed environments; and iii) make the active capability available for heterogeneous sources of events (whether they are databases are not). The objective of the paper is to provide an architecture and framework to support ECA rules for distributed and heterogeneous systems. We describe the design of our ECA rule service, the alternatives considered at each step and the reasons for our choice of a particular alternative. Finally we draw conclusions about the architecture of our system and the utility of our system for a wide range of applications.
主动系统已被提议作为一种范例来满足许多数据库和其他需要及时响应情况的应用程序的需要。事件-条件-动作(ECA)规则(S. Chakravarthy et al.， 1989)用于捕获系统中的活动能力。主动能力(ECA角色)的效用和功能已经在数据库环境中得到了很好的确立。为了使活动功能对大量高级应用程序有用，有必要超越在数据库上下文中提出/开发的内容。具体来说，需要在几个方面对活动能力进行超越当前技术水平的扩展:i)除了数据库应用程序之外，还可以为非数据库应用程序提供活动能力;Ii)使主动能力在分布式环境中可用;iii)使活动功能可用于异构事件源(无论它们是数据库还是非数据库)。本文的目的是提供一个体系结构和框架来支持分布式和异构系统的ECA规则。我们将描述ECA规则服务的设计、每个步骤中考虑的备选方案以及选择特定备选方案的原因。最后，我们对系统的架构和系统的广泛应用做出了总结。
{"title":"ECA rule processing in distributed and heterogeneous environments","authors":"Sharma Chakravarthy, R. Le, R. Dasari","doi":"10.1109/DOA.1999.794056","DOIUrl":"https://doi.org/10.1109/DOA.1999.794056","url":null,"abstract":"Active systems have been proposed as a paradigm to satisfy the needs of many databases and other applications that require a timely response to situations. Event-Condition-Action (or ECA) rules (S. Chakravarthy et al., 1989) are used to capture the active capability in a system. The utility and functionality of active capability (ECA roles) has been well established in the context of databases. In order for the active capability to be useful for a large class of advanced applications, it is necessary to go beyond what has been proposed/developed in the context of databases. Specifically, extensions beyond the current state of the art in active capability are needed along several dimensions: i) make the active capability available for non-database applications, in addition to database applications; ii) make the active capability available in distributed environments; and iii) make the active capability available for heterogeneous sources of events (whether they are databases are not). The objective of the paper is to provide an architecture and framework to support ECA rules for distributed and heterogeneous systems. We describe the design of our ECA rule service, the alternatives considered at each step and the reasons for our choice of a particular alternative. Finally we draw conclusions about the architecture of our system and the utility of our system for a wide range of applications.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128870476","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}
We describe the distributed object oriented architecture of TREVI, a system designed to help overcome the information overload problem. TREVI provides a comprehensive framework for the processing and dissemination of documents coming from news streams and other external sources. TREVI is designed to perform a variety of linguistic and text processing tasks, according to the informational needs expressed in user profiles. Many different user profiles may be simultaneously present, and the sheer volume of information that has to be managed by the system poses serious scalability and performance concerns. We show how the use of a distributed, component based architecture allows us to cope with these challenges, and furthermore provides an excellent base for future system extension, allowing for a rapid response to changing user informational needs.
{"title":"Distributed objects in a large scale text processing system (industrial case study)","authors":"I. Sanz, L. Mazzucchelli","doi":"10.1109/DOA.1999.794033","DOIUrl":"https://doi.org/10.1109/DOA.1999.794033","url":null,"abstract":"We describe the distributed object oriented architecture of TREVI, a system designed to help overcome the information overload problem. TREVI provides a comprehensive framework for the processing and dissemination of documents coming from news streams and other external sources. TREVI is designed to perform a variety of linguistic and text processing tasks, according to the informational needs expressed in user profiles. Many different user profiles may be simultaneously present, and the sheer volume of information that has to be managed by the system poses serious scalability and performance concerns. We show how the use of a distributed, component based architecture allows us to cope with these challenges, and furthermore provides an excellent base for future system extension, allowing for a rapid response to changing user informational needs.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"02 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133621941","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}
The paper builds on the premise that the emerging distributed object infrastructure will logically evolve into a flexible, dynamic data management environment based on a combination of distributed objects, data services and data self-organisation. In order to fulfil this goal, the distributed object community should draw upon the experience accumulated over the years by database researchers and offer its own long-term vision of distributed data processing. The proposed approach is called dispersed data management to emphasise the fact that the conventional, large information sources are divided into smaller, more manageable data entities that are dynamically distributed across computer networks. If that happens it may well be possible to replace the traditional database systems by general, powerful data management architectures based on common data services.
{"title":"The data network: a distributed object view of dynamic data management","authors":"M. Machura","doi":"10.1109/DOA.1999.793985","DOIUrl":"https://doi.org/10.1109/DOA.1999.793985","url":null,"abstract":"The paper builds on the premise that the emerging distributed object infrastructure will logically evolve into a flexible, dynamic data management environment based on a combination of distributed objects, data services and data self-organisation. In order to fulfil this goal, the distributed object community should draw upon the experience accumulated over the years by database researchers and offer its own long-term vision of distributed data processing. The proposed approach is called dispersed data management to emphasise the fact that the conventional, large information sources are divided into smaller, more manageable data entities that are dynamically distributed across computer networks. If that happens it may well be possible to replace the traditional database systems by general, powerful data management architectures based on common data services.","PeriodicalId":360176,"journal":{"name":"Proceedings of the International Symposium on Distributed Objects and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122235051","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：