Researchers widely use speedup, efficiency, and scalability to assess parallel computing performance. These metrics encourage researchers to use any novel technique to design or improve a parallel system, without paying enough attention to the cost increase that such a technique incurs. However, as national-defense applications are downsizing, commercial applications are the dominant users of parallel systems. Customers and vendors are particularly concerned with whether a parallel system can make a profit. Our major goal is to investigate financially justified parallel computing. To evaluate parallel computing's effectiveness, we use a simple profitup metric to measure how performance, cost, and business production affect profit. We focus on investigating the relationship between cost-effective parallel computing and profit-effective parallel computing.
{"title":"Profit-effective parallel computing","authors":"Yong Yan, Xiaodong Zhang","doi":"10.1109/4434.766967","DOIUrl":"https://doi.org/10.1109/4434.766967","url":null,"abstract":"Researchers widely use speedup, efficiency, and scalability to assess parallel computing performance. These metrics encourage researchers to use any novel technique to design or improve a parallel system, without paying enough attention to the cost increase that such a technique incurs. However, as national-defense applications are downsizing, commercial applications are the dominant users of parallel systems. Customers and vendors are particularly concerned with whether a parallel system can make a profit. Our major goal is to investigate financially justified parallel computing. To evaluate parallel computing's effectiveness, we use a simple profitup metric to measure how performance, cost, and business production affect profit. We focus on investigating the relationship between cost-effective parallel computing and profit-effective parallel computing.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116450662","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}
Is the Earth's climate changing, and to what degree are these changes due to human actions? These are the key questions that must be answered before ratification of the Kyoto Protocol. Measurements must be used to evaluate the credibility of supercomputer models of the atmospheric, oceanic, and biotic systems that determine the world's climate. Unlike weather models, climate models deal with the entire atmosphere, ocean, and ice/land-surface system for decades or even centuries. One limiting factor on the use of these models is the availability of computer power to perform the simulations. The adequacy of computer resources and appropriateness of institutional arrangements for climate modeling in the United States was the subject of a National Research Council (NRC) report. Because climate models are not as strongly parallel as many fluid dynamics models, researchers have tended to use vector computers. Researchers are now working to optimize climate models to run on massively parallel computers such as the Cray T3E. Coordination among the federal agencies funding climate modeling has improved greatly. There will not be one national model, but each agency is supporting a few modeling groups and the modeling groups are exchanging modules between the models. Such interchange will allow researchers to choose the best component for each computational approach for each physical component contributing to a model.
{"title":"A changing climate for climate modeling","authors":"D. Lewin","doi":"10.1109/4434.766978","DOIUrl":"https://doi.org/10.1109/4434.766978","url":null,"abstract":"Is the Earth's climate changing, and to what degree are these changes due to human actions? These are the key questions that must be answered before ratification of the Kyoto Protocol. Measurements must be used to evaluate the credibility of supercomputer models of the atmospheric, oceanic, and biotic systems that determine the world's climate. Unlike weather models, climate models deal with the entire atmosphere, ocean, and ice/land-surface system for decades or even centuries. One limiting factor on the use of these models is the availability of computer power to perform the simulations. The adequacy of computer resources and appropriateness of institutional arrangements for climate modeling in the United States was the subject of a National Research Council (NRC) report. Because climate models are not as strongly parallel as many fluid dynamics models, researchers have tended to use vector computers. Researchers are now working to optimize climate models to run on massively parallel computers such as the Cray T3E. Coordination among the federal agencies funding climate modeling has improved greatly. There will not be one national model, but each agency is supporting a few modeling groups and the modeling groups are exchanging modules between the models. Such interchange will allow researchers to choose the best component for each computational approach for each physical component contributing to a model.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122773630","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}
Reflective memory systems are an effective solution to problems raised by message passing in multicomputer environments. The authors provide an overview of existing and emerging RM systems and offer a forecast of future trends.
{"title":"An overview of reflective memory systems","authors":"M. Jovanovic, V. Milutinovic","doi":"10.1109/4434.766965","DOIUrl":"https://doi.org/10.1109/4434.766965","url":null,"abstract":"Reflective memory systems are an effective solution to problems raised by message passing in multicomputer environments. The authors provide an overview of existing and emerging RM systems and offer a forecast of future trends.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130261158","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 ultimate goal of mobile multimedia systems is to assist users all the time and everywhere by providing the right information at the right place in the right manner. With such assistance, users will not have to waste effort in interacting and interfacing with the mobile device, but will be able to dedicate their attention to the actual environment. To provide this assistance, these systems require information about the user's context. One of the most important pieces of context information is the user's location. An efficient location dependent computing system should provide ubiquitous coverage and should be combined with a network infrastructure that provides access anywhere and at any time. The author discusses the application areas where location-dependent multimedia is useful. He looks at the infrastructure necessary of location-specific multimedia applications, including location sensors and location service requirements.
{"title":"Location-dependent multimedia computing","authors":"A. Krikelis","doi":"10.1109/4434.766979","DOIUrl":"https://doi.org/10.1109/4434.766979","url":null,"abstract":"The ultimate goal of mobile multimedia systems is to assist users all the time and everywhere by providing the right information at the right place in the right manner. With such assistance, users will not have to waste effort in interacting and interfacing with the mobile device, but will be able to dedicate their attention to the actual environment. To provide this assistance, these systems require information about the user's context. One of the most important pieces of context information is the user's location. An efficient location dependent computing system should provide ubiquitous coverage and should be combined with a network infrastructure that provides access anywhere and at any time. The author discusses the application areas where location-dependent multimedia is useful. He looks at the infrastructure necessary of location-specific multimedia applications, including location sensors and location service requirements.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126897914","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}
Sun's Jini technology can make a network look like one large computer. The technology will let developers and manufacturers create a whole range of computerized appliances that can instantly connect into a network to share information and services regardless of the underlying operating system or hardware. Once connected, every other computer, device, and user on the network will know that the new device has been added and is available. If Sun succeeds in getting developers to implement and improve Jini, it will have a tremendous impact on networking, especially for wireless and mobile users. Devices in a Jini distributed system provide their own interfaces, which ensures reliability and compatibility. Jini provides the distributed-system services for lookup, registration, and leasing. It consists of four program layers: directory service; JavaSpace; remote-method invocation; and the boot, join, and discover protocol. Any device with an operating system that supports a Java virtual machine can be plugged into the network. The author discusses the origins of Jini, its functions and its competition.
{"title":"Network nirvana and the intelligent device","authors":"D. Clark","doi":"10.1109/4434.766980","DOIUrl":"https://doi.org/10.1109/4434.766980","url":null,"abstract":"Sun's Jini technology can make a network look like one large computer. The technology will let developers and manufacturers create a whole range of computerized appliances that can instantly connect into a network to share information and services regardless of the underlying operating system or hardware. Once connected, every other computer, device, and user on the network will know that the new device has been added and is available. If Sun succeeds in getting developers to implement and improve Jini, it will have a tremendous impact on networking, especially for wireless and mobile users. Devices in a Jini distributed system provide their own interfaces, which ensures reliability and compatibility. Jini provides the distributed-system services for lookup, registration, and leasing. It consists of four program layers: directory service; JavaSpace; remote-method invocation; and the boot, join, and discover protocol. Any device with an operating system that supports a Java virtual machine can be plugged into the network. The author discusses the origins of Jini, its functions and its competition.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132982373","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}
This article presents methods and tools for building parallel applications based on commodity components: PCs, SCSI disks, Fast Ethernet, Windows NT. Chief among these tools is CAP, our computer-aided parallelization tool. CAP generates highly pipelined applications that run communication and I/O operations in parallel with processing operations. One of CAP's successes is the Visible Human Slice Server, a 3D tomographic image server that allows clients to choose and view any cross section of the human body.
{"title":"Parallelizing I/O-intensive image access and processing applications","authors":"V. Messerli, O. Figueiredo, B. Gennart, R. Hersch","doi":"10.1109/4434.766960","DOIUrl":"https://doi.org/10.1109/4434.766960","url":null,"abstract":"This article presents methods and tools for building parallel applications based on commodity components: PCs, SCSI disks, Fast Ethernet, Windows NT. Chief among these tools is CAP, our computer-aided parallelization tool. CAP generates highly pipelined applications that run communication and I/O operations in parallel with processing operations. One of CAP's successes is the Visible Human Slice Server, a 3D tomographic image server that allows clients to choose and view any cross section of the human body.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124254295","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}
Software parallelization is required to contend with the increasing scale and complexity of high-energy physics experiments. The authors have developed a programming model, Communication Capability (CoCa) which allows this parallelization at several levels of granularity and reduces software complexity.
{"title":"CoCa: a parallelization model for high energy physics","authors":"P. V. D. Stok, E. Argante, I. Willers","doi":"10.1109/4434.766962","DOIUrl":"https://doi.org/10.1109/4434.766962","url":null,"abstract":"Software parallelization is required to contend with the increasing scale and complexity of high-energy physics experiments. The authors have developed a programming model, Communication Capability (CoCa) which allows this parallelization at several levels of granularity and reduces software complexity.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116831682","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}
S. Russ, Katina Reece, Jonathan Robinson, B. Meyers, D. Rajan, L. Rajagopalan, Chun-Heong Tan
The authors present the Hector distributed runtime environment, in which "slave allocators" forward commands from the central allocation system, monitor system availability and loading, and can transparently extract detailed performance information from programs as they run. The authors detail the overall architecture of the agent-based allocation system and demonstrate its efficiency, low overhead, and utility.
{"title":"Hector: an agent based architecture for dynamic resource management","authors":"S. Russ, Katina Reece, Jonathan Robinson, B. Meyers, D. Rajan, L. Rajagopalan, Chun-Heong Tan","doi":"10.1109/4434.766964","DOIUrl":"https://doi.org/10.1109/4434.766964","url":null,"abstract":"The authors present the Hector distributed runtime environment, in which \"slave allocators\" forward commands from the central allocation system, monitor system availability and loading, and can transparently extract detailed performance information from programs as they run. The authors detail the overall architecture of the agent-based allocation system and demonstrate its efficiency, low overhead, and utility.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125044881","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 author introduces smart cards, discussing both their background and where they're headed in the future. The term smart card generally refers to a plastic card the size of an ordinary credit card with a chip that holds a microprocessor and a data-storage unit. Developers hope smart cards will combine conventional identification documents with an electronic purse and a bank account access card for telebanking transactions. Another goal is to be able to insert the cards into an Internet terminal for e-business. Smart cards might also provide access to mail functions; replace key chains by acting as a multipurpose digital key; serve as a ticket for public transportation systems; or serve as a key card for a previously booked hotel room. The author discusses the history of smart cards, standards, communication, and their use in mobile applications.
{"title":"The smart card: don't leave home without it","authors":"D. Husemann","doi":"10.1109/4434.766959","DOIUrl":"https://doi.org/10.1109/4434.766959","url":null,"abstract":"The author introduces smart cards, discussing both their background and where they're headed in the future. The term smart card generally refers to a plastic card the size of an ordinary credit card with a chip that holds a microprocessor and a data-storage unit. Developers hope smart cards will combine conventional identification documents with an electronic purse and a bank account access card for telebanking transactions. Another goal is to be able to insert the cards into an Internet terminal for e-business. Smart cards might also provide access to mail functions; replace key chains by acting as a multipurpose digital key; serve as a ticket for public transportation systems; or serve as a key card for a previously booked hotel room. The author discusses the history of smart cards, standards, communication, and their use in mobile applications.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"240 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130986044","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}
Ada 83 has numerous shortcomings for concurrent systems: it requires additional tasks, adding overhead and increasing interrupt latency, and it has difficulty handling asynchronous transfer of control. Ada 95 offers features that address these shortcomings and simplify concurrent implementation: protected units, interrupt handlers, priority entry queuing, dynamic task priorities, and asynchronous select statements. Using their Flexible Manufacturing System, the authors demonstrate how these features improve Ada and promote efficiency.
Ada 83对于并发系统有许多缺点:它需要额外的任务、增加开销和增加中断延迟，并且难以处理异步控制传输。Ada 95提供了解决这些缺点和简化并发实现的特性:受保护单元、中断处理程序、优先级条目排队、动态任务优先级和异步选择语句。利用他们的柔性制造系统，作者演示了这些特性如何改善Ada和提高效率。
{"title":"Practical use of Ada 95's concurrency features","authors":"J. Carter, B. Sandén","doi":"10.1109/4434.736425","DOIUrl":"https://doi.org/10.1109/4434.736425","url":null,"abstract":"Ada 83 has numerous shortcomings for concurrent systems: it requires additional tasks, adding overhead and increasing interrupt latency, and it has difficulty handling asynchronous transfer of control. Ada 95 offers features that address these shortcomings and simplify concurrent implementation: protected units, interrupt handlers, priority entry queuing, dynamic task priorities, and asynchronous select statements. Using their Flexible Manufacturing System, the authors demonstrate how these features improve Ada and promote efficiency.","PeriodicalId":282630,"journal":{"name":"IEEE Concurr.","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125125191","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：