Pub Date : 1991-12-01DOI: 10.1109/PDIS.1991.183084
I. J. Aalbersberg, F. Sijstermans
Describes the InfoGuide full-text document retrieval system, developed at the Philips Research Laboratories. InfoGuide combines high retrieval quality with high retrieval performance. The high quality is due to the use of the vector-space retrieval model, and the high performance originates from the implementation on the parallel POOMA machine. Implementation aspects as well as evaluation figures are presented, and it is concluded that a 200-term query can be matched against 1,500,000 documents within 3 seconds.<>
{"title":"High-quality and high-performance full-text document retrieval: the Parallel InfoGuide System","authors":"I. J. Aalbersberg, F. Sijstermans","doi":"10.1109/PDIS.1991.183084","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183084","url":null,"abstract":"Describes the InfoGuide full-text document retrieval system, developed at the Philips Research Laboratories. InfoGuide combines high retrieval quality with high retrieval performance. The high quality is due to the use of the vector-space retrieval model, and the high performance originates from the implementation on the parallel POOMA machine. Implementation aspects as well as evaluation figures are presented, and it is concluded that a 200-term query can be matched against 1,500,000 documents within 3 seconds.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114002865","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183109
David M. Choy, P. Selinger
To support a distributed, heterogeneous computing environment, an inter-system catalog protocol is needed so that remote resources can be located, used, and maintained with little human intervention. This paper describes a scalable catalog framework, which is an extension of previous work in a distributed relational DBMS research prototype called R*. This work builds on the R* concepts to accommodate heterogeneity, to handle partitioned and replicated data, to support non-DBMS resource managers, and to enhance catalog access performance and system extensibility.<>
{"title":"A distributed catalog for heterogeneous distributed database resources","authors":"David M. Choy, P. Selinger","doi":"10.1109/PDIS.1991.183109","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183109","url":null,"abstract":"To support a distributed, heterogeneous computing environment, an inter-system catalog protocol is needed so that remote resources can be located, used, and maintained with little human intervention. This paper describes a scalable catalog framework, which is an extension of previous work in a distributed relational DBMS research prototype called R*. This work builds on the R* concepts to accommodate heterogeneity, to handle partitioned and replicated data, to support non-DBMS resource managers, and to enhance catalog access performance and system extensibility.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122917672","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183096
A. Grimshaw, E. Loyot
Summary form only given. ELFS (an ExtensibLe File System), (1) provides high bandwidth and low latency I/O to applications programs on high performance architectures, (2) reduces the cognitive burden faced by applications programmers when they attempt to optimize their I/O operations to fit existing file system models, and (3) seamlessly manages the proliferation of data formats and architectural differences. The ELFS solution consists of language and run-time system support that permits the specification of a hierarchy of file classes.<>
{"title":"ELFS: object-oriented extensible file systems","authors":"A. Grimshaw, E. Loyot","doi":"10.1109/PDIS.1991.183096","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183096","url":null,"abstract":"Summary form only given. ELFS (an ExtensibLe File System), (1) provides high bandwidth and low latency I/O to applications programs on high performance architectures, (2) reduces the cognitive burden faced by applications programmers when they attempt to optimize their I/O operations to fit existing file system models, and (3) seamlessly manages the proliferation of data formats and architectural differences. The ELFS solution consists of language and run-time system support that permits the specification of a hierarchy of file classes.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124081198","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183070
J. Wolf, D. Dias, Philip S. Yu, John Turek
The authors recently (1990, 1991) described two new join algorithms designed to address the data skew problem. These algorithms were based, respectively, on the traditional sort merge and hash join algorithms, and employed techniques borrowed from mathematical optimization theory. The current paper proposes significant improvements to both algorithms, increasing their effectiveness while simultaneously decreasing their execution times. It then focuses on the comparative performance of the improved algorithms and their more conventional sort merge and hash counterparts. The latter two are perfectly good algorithms except that they fail to deal with data skew. Both I/O- and CPU-bound configurations were examined. The new algorithms outperform their more conventional counterparts in the presence of just about any skew at all, dramatically so in cases of high skew.<>
{"title":"Comparative performance of parallel join algorithms","authors":"J. Wolf, D. Dias, Philip S. Yu, John Turek","doi":"10.1109/PDIS.1991.183070","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183070","url":null,"abstract":"The authors recently (1990, 1991) described two new join algorithms designed to address the data skew problem. These algorithms were based, respectively, on the traditional sort merge and hash join algorithms, and employed techniques borrowed from mathematical optimization theory. The current paper proposes significant improvements to both algorithms, increasing their effectiveness while simultaneously decreasing their execution times. It then focuses on the comparative performance of the improved algorithms and their more conventional sort merge and hash counterparts. The latter two are perfectly good algorithms except that they fail to deal with data skew. Both I/O- and CPU-bound configurations were examined. The new algorithms outperform their more conventional counterparts in the presence of just about any skew at all, dramatically so in cases of high skew.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115755311","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183111
Ming-Syan Chen, Philip S. Yu, Kun-Lung Wu
Broadcasting, which refers to a process of information dissemination in a distributed system whereby a message originating from a certain node is sent to all other nodes in the system, is a very important issue in distributed computing. All-to-all broadcasting means the process by which every node broadcasts its certain piece of information to all other nodes. The authors develop optimal all-to-all broadcasting schemes for a distributed system of an arbitrary number of nodes to complete the broadcasting with not only the minimal number of communication steps but also the minimal number of messages. They develop the optimal all-to-all broadcasting scheme for the case of k-port communication, meaning that each node can send out k messages in one communication step where k is a positive integer depending on the system. It is shown that the proposed scheme not only requires the minimal number of communication steps but also incurs the minimal number of messages.<>
{"title":"Optimal all-to-all broadcasting schemes in distributed systems","authors":"Ming-Syan Chen, Philip S. Yu, Kun-Lung Wu","doi":"10.1109/PDIS.1991.183111","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183111","url":null,"abstract":"Broadcasting, which refers to a process of information dissemination in a distributed system whereby a message originating from a certain node is sent to all other nodes in the system, is a very important issue in distributed computing. All-to-all broadcasting means the process by which every node broadcasts its certain piece of information to all other nodes. The authors develop optimal all-to-all broadcasting schemes for a distributed system of an arbitrary number of nodes to complete the broadcasting with not only the minimal number of communication steps but also the minimal number of messages. They develop the optimal all-to-all broadcasting scheme for the case of k-port communication, meaning that each node can send out k messages in one communication step where k is a positive integer depending on the system. It is shown that the proposed scheme not only requires the minimal number of communication steps but also incurs the minimal number of messages.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131989852","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183073
Weining Zhang, Ke Wang, Siu-Cheung Chau
The authors analyse previous parallel evaluation strategies based on the partition of rule instantiations. They present a parallel evaluation strategy for general Datalog programs that is based on the partition of data and is more practical. A key issue is to determine a criterion of data transmission that reduces the amount of data transmitted and is tested efficiently. A notion of potential usefulness is given as such a criterion. The problem of designing appropriate partition schemes and processing schemes is addressed. Heuristics and algorithms are proposed for making decisions in the design process.<>
{"title":"Data partition: a practical parallel evaluation of datalog programs","authors":"Weining Zhang, Ke Wang, Siu-Cheung Chau","doi":"10.1109/PDIS.1991.183073","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183073","url":null,"abstract":"The authors analyse previous parallel evaluation strategies based on the partition of rule instantiations. They present a parallel evaluation strategy for general Datalog programs that is based on the partition of data and is more practical. A key issue is to determine a criterion of data transmission that reduces the amount of data transmitted and is tested efficiently. A notion of potential usefulness is given as such a criterion. The problem of designing appropriate partition schemes and processing schemes is addressed. Heuristics and algorithms are proposed for making decisions in the design process.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130347969","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183097
T. Bowen, M. Cochinwala, G. Gopal, G. Herman, T. Hickey, Kuo Chi Lee, W. Mansfield, J. Raitz
Summary form only given. The Datacycle database architecture is based on repetitive broadcast of database contents over high bandwidth channels and on-the-fly filtering of database contents to extract records of interest. These two mechanisms-high speed broadcast and on-the-fly filtering-result in a system that achieves the goals of high performance transaction processing, a powerful and flexible query capability, and high levels of concurrent access to a single database by multiple applications. The prototype system comprises multiple pump boards, each of which stores up to 128 Mbytes of database contents, and multiple access manager boards, each of which contains three on-board VLSI data filter chips. Each VLSI filter executes queries against the contents of a 53 Mbyte/second broadcast channel; selected records and/or computed aggregate values are returned to the host application.<>
{"title":"Achieving throughput and functionality in a common architecture: the Datacycle experiment","authors":"T. Bowen, M. Cochinwala, G. Gopal, G. Herman, T. Hickey, Kuo Chi Lee, W. Mansfield, J. Raitz","doi":"10.1109/PDIS.1991.183097","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183097","url":null,"abstract":"Summary form only given. The Datacycle database architecture is based on repetitive broadcast of database contents over high bandwidth channels and on-the-fly filtering of database contents to extract records of interest. These two mechanisms-high speed broadcast and on-the-fly filtering-result in a system that achieves the goals of high performance transaction processing, a powerful and flexible query capability, and high levels of concurrent access to a single database by multiple applications. The prototype system comprises multiple pump boards, each of which stores up to 128 Mbytes of database contents, and multiple access manager boards, each of which contains three on-board VLSI data filter chips. Each VLSI filter executes queries against the contents of a 53 Mbyte/second broadcast channel; selected records and/or computed aggregate values are returned to the host application.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114274621","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183110
S. Mehrotra, R. Rastogi, H. F. Korth, A. Silberschatz
The concept of serializability has been the traditionally accepted notion of correctness in database systems. However, in a heterogeneous distributed database system (HDBMS) environment, ensuring serializability is a difficult task mainly due to the desire of preserving the local autonomy of the participating local database systems. The authors introduce a new correctness criterion fur HDBMSs, two-level serializability (2LSR), and show that it preserves database consistency in several HDBMS models. Further, they present a simple protocol for ensuring that schedules in an HDBMS environment are 2LSR. This protocol is easily implementable and does not violate the local autonomy of sites.<>
{"title":"Non-serializable executions in heterogeneous distributed database systems","authors":"S. Mehrotra, R. Rastogi, H. F. Korth, A. Silberschatz","doi":"10.1109/PDIS.1991.183110","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183110","url":null,"abstract":"The concept of serializability has been the traditionally accepted notion of correctness in database systems. However, in a heterogeneous distributed database system (HDBMS) environment, ensuring serializability is a difficult task mainly due to the desire of preserving the local autonomy of the participating local database systems. The authors introduce a new correctness criterion fur HDBMSs, two-level serializability (2LSR), and show that it preserves database consistency in several HDBMS models. Further, they present a simple protocol for ensuring that schedules in an HDBMS environment are 2LSR. This protocol is easily implementable and does not violate the local autonomy of sites.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123905029","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183081
B. Hillyer, B. Robinson
Summary form only given. As a consequence of improving technology, a new class of applications is emerging, characterized by intense demands on communications, computation, and data storage. BBFS is a broadband filesystem effort to explore ways of supporting these applications. A goal of BBFS is to simplify applications by absorbing data-intensive processing into the filesystem. To obtain behavioral extensibility, BBFS provides the notion of file types that encapsulate data structures and behaviors, such as structured files, version history files, spanned files that cross device boundaries, mirrored files, and active files containing data with associated functions.<>
{"title":"Aspects of the BBFS broadband filesystem","authors":"B. Hillyer, B. Robinson","doi":"10.1109/PDIS.1991.183081","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183081","url":null,"abstract":"Summary form only given. As a consequence of improving technology, a new class of applications is emerging, characterized by intense demands on communications, computation, and data storage. BBFS is a broadband filesystem effort to explore ways of supporting these applications. A goal of BBFS is to simplify applications by absorbing data-intensive processing into the filesystem. To obtain behavioral extensibility, BBFS provides the notion of file types that encapsulate data structures and behaviors, such as structured files, version history files, spanned files that cross device boundaries, mirrored files, and active files containing data with associated functions.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123988872","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 : 1991-12-01DOI: 10.1109/PDIS.1991.183091
Yasushi Shinjo, Y. Kiyoki
Summary form only. ReSC is a distributed operating system for both parallel/distributed applications and sequential applications. The hardware environment of ReSC is a set of shared-memory multiprocessors and uniprocessors which are connected to a high-speed local area network. To execute parallel/distributed application programs efficiently, those programs should control scheduling and mapping of their processes by themselves. ReSC provides facilities that enable each application program to control scheduling and mapping. The authors refer to the module which controls mapping as a distribution optimizer. In ReSC, each parallel/distributed application program can include its own lightweight process scheduler and distribution optimiser for optimal mapping and scheduling. ReSC supports for them by kernel calls and libraries. They have developed a lightweight process library on Sequent Balance 8000 and Omron Luna88k (which are shared-memory multiprocessors), Sun4, Sun3 and NeXT. By using the library, they have built a parallel processing system for databases.<>
{"title":"ReSC: a distributed operating system for parallel and distributed applications","authors":"Yasushi Shinjo, Y. Kiyoki","doi":"10.1109/PDIS.1991.183091","DOIUrl":"https://doi.org/10.1109/PDIS.1991.183091","url":null,"abstract":"Summary form only. ReSC is a distributed operating system for both parallel/distributed applications and sequential applications. The hardware environment of ReSC is a set of shared-memory multiprocessors and uniprocessors which are connected to a high-speed local area network. To execute parallel/distributed application programs efficiently, those programs should control scheduling and mapping of their processes by themselves. ReSC provides facilities that enable each application program to control scheduling and mapping. The authors refer to the module which controls mapping as a distribution optimizer. In ReSC, each parallel/distributed application program can include its own lightweight process scheduler and distribution optimiser for optimal mapping and scheduling. ReSC supports for them by kernel calls and libraries. They have developed a lightweight process library on Sequent Balance 8000 and Omron Luna88k (which are shared-memory multiprocessors), Sun4, Sun3 and NeXT. By using the library, they have built a parallel processing system for databases.<<ETX>>","PeriodicalId":210800,"journal":{"name":"[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1991-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115816343","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：