The heterogeneous parallel processing community has long been struggling to bring its approach to computation into the mainstream. One major impediment is that no popular programming language supports a sufficiently wide range of models of parallelism. The recent emergence of Java as a popular programming language may offer an opportunity to change this situation. The article begins with a review of the special linguistic and computational needs of heterogeneous parallel processing by considering the user communities that would benefit most from the approach. It then reviews the pros and cons of Java as a language for expressing and realizing heterogeneity, and concludes with some possible changes that would make Java more suitable for such use.
{"title":"Heterogeneous programming with Java: gourmet blend or just a hill of beans?","authors":"C. Weems","doi":"10.1109/HCW.1998.666557","DOIUrl":"https://doi.org/10.1109/HCW.1998.666557","url":null,"abstract":"The heterogeneous parallel processing community has long been struggling to bring its approach to computation into the mainstream. One major impediment is that no popular programming language supports a sufficiently wide range of models of parallelism. The recent emergence of Java as a popular programming language may offer an opportunity to change this situation. The article begins with a review of the special linguistic and computational needs of heterogeneous parallel processing by considering the user communities that would benefit most from the approach. It then reviews the pros and cons of Java as a language for expressing and realizing heterogeneity, and concludes with some possible changes that would make Java more suitable for such use.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127543861","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. F. Freund, M. Gherrity, Stephen L. Ambrosius, M. Campbell, Mike Halderman, D. Hensgen, E. Keith, T. Kidd, M. Kussow, J. Lima, F. Mirabile, L. Moore, B. Rust, H. Siegel
It is increasingly common for computer users to have access to several computers on a network, and hence to be able to execute many of their tasks on any of several computers. The choice of which computers execute which tasks is commonly determined by users based on a knowledge of computer speeds for each task and the current load on each computer. A number of task scheduling systems have been developed that balance the load of the computers on the network, but such systems tend to minimize the idle time of the computers rather than minimize the idle time of the users. The paper focuses on the benefits that can be achieved when the scheduling system considers both the computer availabilities and the performance of each task on each computer. The SmartNet resource scheduling system is described and compared to two different resource allocation strategies: load balancing and user directed assignment. Results are presented where the operation of hundreds of different networks of computers running thousands of different mixes of tasks are simulated in a batch environment. These results indicate that, for the computer environments simulated, SmartNet outperforms both load balancing and user directed assignments, based on the maximum time users must wait for their tasks to finish.
{"title":"Scheduling resources in multi-user, heterogeneous, computing environments with SmartNet","authors":"R. F. Freund, M. Gherrity, Stephen L. Ambrosius, M. Campbell, Mike Halderman, D. Hensgen, E. Keith, T. Kidd, M. Kussow, J. Lima, F. Mirabile, L. Moore, B. Rust, H. Siegel","doi":"10.1109/HCW.1998.666558","DOIUrl":"https://doi.org/10.1109/HCW.1998.666558","url":null,"abstract":"It is increasingly common for computer users to have access to several computers on a network, and hence to be able to execute many of their tasks on any of several computers. The choice of which computers execute which tasks is commonly determined by users based on a knowledge of computer speeds for each task and the current load on each computer. A number of task scheduling systems have been developed that balance the load of the computers on the network, but such systems tend to minimize the idle time of the computers rather than minimize the idle time of the users. The paper focuses on the benefits that can be achieved when the scheduling system considers both the computer availabilities and the performance of each task on each computer. The SmartNet resource scheduling system is described and compared to two different resource allocation strategies: load balancing and user directed assignment. Results are presented where the operation of hundreds of different networks of computers running thousands of different mixes of tasks are simulated in a batch environment. These results indicate that, for the computer environments simulated, SmartNet outperforms both load balancing and user directed assignments, based on the maximum time users must wait for their tasks to finish.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"488 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116115559","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 Java a good language for programming heterogeneous parallel computing systems? It is a well-designed modern language that, combined with the Java Virtual Machine (JVM), offers a myriad of modern programming features and excellent portability. However, in speedup-oriented heterogeneous computing, the primary concern is obtaining the best possible execution speed from the heterogeneous system. The paper briefly discusses what heterogeneous parallel computing is really about, lists some of the key features of Java, and finally summarizes how well Java matches the task of programming for heterogeneous parallel computing.
{"title":"Heterogeneous parallel computing with Java: jabber or justified?","authors":"H. Dietz","doi":"10.1109/HCW.1998.666554","DOIUrl":"https://doi.org/10.1109/HCW.1998.666554","url":null,"abstract":"Is Java a good language for programming heterogeneous parallel computing systems? It is a well-designed modern language that, combined with the Java Virtual Machine (JVM), offers a myriad of modern programming features and excellent portability. However, in speedup-oriented heterogeneous computing, the primary concern is obtaining the best possible execution speed from the heterogeneous system. The paper briefly discusses what heterogeneous parallel computing is really about, lists some of the key features of Java, and finally summarizes how well Java matches the task of programming for heterogeneous parallel computing.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130625570","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 Globus project is a multi-institutional research effort that seeks to enable the construction of computational grids providing pervasive, dependable, and consistent access to high-performance computational resources, despite geographical distribution of both resources and users. Computational grid technology is being viewed as a critical element of future high-performance computing environments that will enable entirely new classes of computation-oriented applications, much as the World Wide Web fostered the development of new classes of information-oriented applications. The authors report on the status of the Globus project as of early 1998. They describe the progress that has been achieved to date in the development of the Globus toolkit, a set of core services for constructing grid tools and applications. They also discuss the Globus Ubiquitous Supercomputing Testbed (GUSTO) that they have constructed to enable large-scale evaluation of Globus technologies, and they review early experiences with the development of large-scale grid applications on the GUSTO testbed.
{"title":"The Globus project: a status report","authors":"Ian T Foster, C. Kesselman","doi":"10.1109/HCW.1998.666541","DOIUrl":"https://doi.org/10.1109/HCW.1998.666541","url":null,"abstract":"The Globus project is a multi-institutional research effort that seeks to enable the construction of computational grids providing pervasive, dependable, and consistent access to high-performance computational resources, despite geographical distribution of both resources and users. Computational grid technology is being viewed as a critical element of future high-performance computing environments that will enable entirely new classes of computation-oriented applications, much as the World Wide Web fostered the development of new classes of information-oriented applications. The authors report on the status of the Globus project as of early 1998. They describe the progress that has been achieved to date in the development of the Globus toolkit, a set of core services for constructing grid tools and applications. They also discuss the Globus Ubiquitous Supercomputing Testbed (GUSTO) that they have constructed to enable large-scale evaluation of Globus technologies, and they review early experiences with the development of large-scale grid applications on the GUSTO testbed.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121232004","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}
Data staging is an important data management problem for a distributed heterogeneous networking environment, where each data storage location and intermediate node may have specific data available, storage limitations, and communication links. Sites in the network request data items and each item is associated with a specific deadline and priority. It is assumed that not all requests can be satisfied by their deadline. The work concentrates on solving a basic version of the data staging problem in which all parameter values for the communication system and the data request information represent the best known information collected so far and stay fixed throughout the scheduling process. A mathematical model for the basic data staging problem is introduced. Then, a multiple-source shortest-path algorithm based heuristic for finding a suboptimal schedule of the communication steps for data staging is presented. A simulation study is provided, which evaluates the performance of the proposed heuristic. The results show the advantages of the proposed heuristic over two random based scheduling techniques. This research, based on the simplified static model, serves as a necessary step toward solving the more realistic and complicated version of the data staging problem involving dynamic scheduling, fault tolerance, and determining where to stage data.
{"title":"A mathematical model, heuristic, and simulation study for a basic data staging problem in a heterogeneous networking environment","authors":"Min Tan, M. Theys, H. Siegel, N. Beck, M. Jurczyk","doi":"10.1109/HCW.1998.666550","DOIUrl":"https://doi.org/10.1109/HCW.1998.666550","url":null,"abstract":"Data staging is an important data management problem for a distributed heterogeneous networking environment, where each data storage location and intermediate node may have specific data available, storage limitations, and communication links. Sites in the network request data items and each item is associated with a specific deadline and priority. It is assumed that not all requests can be satisfied by their deadline. The work concentrates on solving a basic version of the data staging problem in which all parameter values for the communication system and the data request information represent the best known information collected so far and stay fixed throughout the scheduling process. A mathematical model for the basic data staging problem is introduced. Then, a multiple-source shortest-path algorithm based heuristic for finding a suboptimal schedule of the communication steps for data staging is presented. A simulation study is provided, which evaluates the performance of the proposed heuristic. The results show the advantages of the proposed heuristic over two random based scheduling techniques. This research, based on the simplified static model, serves as a necessary step toward solving the more realistic and complicated version of the data staging problem involving dynamic scheduling, fault tolerance, and determining where to stage data.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124148465","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 NetSolve project, underway at the University of Tennessee and at the Oak Ridge National Laboratory, allows users to access computational resources distributed across the network. These resources are embodied in computational servers and allow the user to easily perform scientific computing tasks without having any computing facility installed on his/her computer. The user access to the servers is facilitated by a variety of interfaces: application programming interfaces (APIs), textual interactive interfaces and graphical user interfaces (GUIs). There are many research issues involved in the NetSolve system including fault-tolerance, load balancing, user-interface design, computational servers and network-based computing. As the project matures several promising extensions and applications of NetSolve will emerge. The authors provide an overview of the project and examine some of the extensions being developed: an interface to the Condor system, can interface to the ScaLAPACK parallel library, a bridge with the Ninf system, and an integration of NetSolve and Image Vision.
{"title":"NetSolve: a network-enabled solver; examples and users","authors":"H. Casanova, J. Dongarra","doi":"10.1109/HCW.1998.666542","DOIUrl":"https://doi.org/10.1109/HCW.1998.666542","url":null,"abstract":"The NetSolve project, underway at the University of Tennessee and at the Oak Ridge National Laboratory, allows users to access computational resources distributed across the network. These resources are embodied in computational servers and allow the user to easily perform scientific computing tasks without having any computing facility installed on his/her computer. The user access to the servers is facilitated by a variety of interfaces: application programming interfaces (APIs), textual interactive interfaces and graphical user interfaces (GUIs). There are many research issues involved in the NetSolve system including fault-tolerance, load balancing, user-interface design, computational servers and network-based computing. As the project matures several promising extensions and applications of NetSolve will emerge. The authors provide an overview of the project and examine some of the extensions being developed: an interface to the Condor system, can interface to the ScaLAPACK parallel library, a bridge with the Ninf system, and an integration of NetSolve and Image Vision.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"283 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133003243","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}
G. Agha, Mark Astley, Jamil A. Sheikh, Carlos A. Varela
Java supports heterogeneous applications by transforming a heterogeneous network of machines into a homogeneous network of Java virtual machines. This approach abstracts over many of the complications that arise from heterogeneity, providing a uniform API to all components of an application. However, for many applications heterogeneity is an intentional feature where components and resources are co-located for optimal performance. The authors argue that Java's API does not provide an effective means for building applications in such an environment. Specifically, they suggest improvements to Java's existing mechanisms for maintaining consistency (e.g. synchronized), and controlling resources (e.g. thread scheduling). They also consider the recent addition of a CORBA API in JDK 1.2. They argue that while such an approach provides greater flexibility for heterogeneous applications, many key problems still exist from an architectural standpoint. Finally, they consider the future of Java as a foundation for component-based software in heterogeneous environments and suggest architectural abstractions which will prove key to the successful development of such systems. They drive the discussion with examples and suggestions from their work on the Actor model of computation.
{"title":"Modular heterogeneous system development: a critical analysis of Java","authors":"G. Agha, Mark Astley, Jamil A. Sheikh, Carlos A. Varela","doi":"10.1109/HCW.1998.666552","DOIUrl":"https://doi.org/10.1109/HCW.1998.666552","url":null,"abstract":"Java supports heterogeneous applications by transforming a heterogeneous network of machines into a homogeneous network of Java virtual machines. This approach abstracts over many of the complications that arise from heterogeneity, providing a uniform API to all components of an application. However, for many applications heterogeneity is an intentional feature where components and resources are co-located for optimal performance. The authors argue that Java's API does not provide an effective means for building applications in such an environment. Specifically, they suggest improvements to Java's existing mechanisms for maintaining consistency (e.g. synchronized), and controlling resources (e.g. thread scheduling). They also consider the recent addition of a CORBA API in JDK 1.2. They argue that while such an approach provides greater flexibility for heterogeneous applications, many key problems still exist from an architectural standpoint. Finally, they consider the future of Java as a foundation for component-based software in heterogeneous environments and suggest architectural abstractions which will prove key to the successful development of such systems. They drive the discussion with examples and suggestions from their work on the Actor model of computation.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122339507","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}
Java's design goals of portability, safety, and ubiquity make it a potentially ideal language for large-scale heterogeneous computing. One of the remaining challenges is to create performance models and associated specifications and programming constructs that can be used to reason about performance properties of systems implemented in Java.
{"title":"Steps toward understanding performance in Java","authors":"D. Lea","doi":"10.1109/HCW.1998.666556","DOIUrl":"https://doi.org/10.1109/HCW.1998.666556","url":null,"abstract":"Java's design goals of portability, safety, and ubiquity make it a potentially ideal language for large-scale heterogeneous computing. One of the remaining challenges is to create performance models and associated specifications and programming constructs that can be used to reason about performance properties of systems implemented in Java.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132791102","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}
CCS is a resource management system for parallel high-performance computers. At the user level, CCS provides vendor-independent access to parallel systems. At the system administrator level, CCS offers tools for controlling (i.e, specifying, configuring and scheduling) the system components that are operated in a computing center. Hence the name "Computing Center Software". CCS provides: hardware-independent scheduling of interactive and batch jobs; partitioning of exclusive and non-exclusive resources; open, extensible interfaces to other resource management systems; a high degree of reliability (e.g. automatic restart of crashed daemons); fault tolerance in the case of network breakdowns. The authors describe CCS as one important component for the access, job distribution, and administration of networked HPC systems in a metacomputing environment.
{"title":"CCS resource management in networked HPC systems","authors":"A. Keller, A. Reinefeld","doi":"10.1109/HCW.1998.666544","DOIUrl":"https://doi.org/10.1109/HCW.1998.666544","url":null,"abstract":"CCS is a resource management system for parallel high-performance computers. At the user level, CCS provides vendor-independent access to parallel systems. At the system administrator level, CCS offers tools for controlling (i.e, specifying, configuring and scheduling) the system components that are operated in a computing center. Hence the name \"Computing Center Software\". CCS provides: hardware-independent scheduling of interactive and batch jobs; partitioning of exclusive and non-exclusive resources; open, extensible interfaces to other resource management systems; a high degree of reliability (e.g. automatic restart of crashed daemons); fault tolerance in the case of network breakdowns. The authors describe CCS as one important component for the access, job distribution, and administration of networked HPC systems in a metacomputing environment.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117295722","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}
Java has been described as a simple, object-oriented, distributed, interpreted, robust, secure, architectural neutral, portable, high-performance, multithreaded and dynamic language, prompting some to describe it as the first buzzword-compliant programming language. The author submits that to deserve full certification-and in the process establish itself as the natural choice for developing large-scale distributed applications-Java misses a crucial buzzword: fault-tolerant. He outlines some promising research directions for building reliable Java-based applications.
{"title":"Fault-tolerance: Java's missing buzzword","authors":"L. Alvisi","doi":"10.1109/HCW.1998.666553","DOIUrl":"https://doi.org/10.1109/HCW.1998.666553","url":null,"abstract":"Java has been described as a simple, object-oriented, distributed, interpreted, robust, secure, architectural neutral, portable, high-performance, multithreaded and dynamic language, prompting some to describe it as the first buzzword-compliant programming language. The author submits that to deserve full certification-and in the process establish itself as the natural choice for developing large-scale distributed applications-Java misses a crucial buzzword: fault-tolerant. He outlines some promising research directions for building reliable Java-based applications.","PeriodicalId":273718,"journal":{"name":"Proceedings Seventh Heterogeneous Computing Workshop (HCW'98)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132146957","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：