Pub Date : 1992-01-07DOI: 10.1109/HICSS.1992.183147
J. Menon, J. Kasson
A disk array is a set of disk drives (and controller) which can automatically recover data when one (or more) disk drives in the set fails. One method used by disk arrays to achieve high availability at lower cost than mirroring is a parity technique. The main drawback of such arrays are that they need four disk accesses to update a data block-two to read old data and parity, and two to write new data and parity. The authors describe four new methods to improve the update performance of disk arrays that use the parity technique from four accesses to three and, in some cases, to two. All the schemes sacrifice disk storage efficiency for improved update performance by relaxing the requirement that the modified data and parity blocks be written back into their original locations. The best technique, called 'floating parity track', achieves much improved update performance while using only 1% more disk space than traditional arrays.<>
{"title":"Methods for improved update performance of disk arrays","authors":"J. Menon, J. Kasson","doi":"10.1109/HICSS.1992.183147","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183147","url":null,"abstract":"A disk array is a set of disk drives (and controller) which can automatically recover data when one (or more) disk drives in the set fails. One method used by disk arrays to achieve high availability at lower cost than mirroring is a parity technique. The main drawback of such arrays are that they need four disk accesses to update a data block-two to read old data and parity, and two to write new data and parity. The authors describe four new methods to improve the update performance of disk arrays that use the parity technique from four accesses to three and, in some cases, to two. All the schemes sacrifice disk storage efficiency for improved update performance by relaxing the requirement that the modified data and parity blocks be written back into their original locations. The best technique, called 'floating parity track', achieves much improved update performance while using only 1% more disk space than traditional arrays.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"19 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123033656","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183149
A. Hác
Studies the sensitivity of asynchronous algorithms in disk buffer cache memory. The algorithms allow for writing the files into the buffer cache by the processes. The number of active processes in the system and the length of the queue to the disk buffer cache are considered in sensitivity study. This information is obtained dynamically during the execution of the algorithms. The time for a block to become available for updating in the buffer cache is given. The number of blocks available for updating in the buffer cache is derived. The performance of the algorithms is compared. It is shown that the proposed algorithms allow for better performance than an algorithm that does not use information about the system load.<>
{"title":"Sensitivity study of asynchronous algorithms in disk buffer cache memory","authors":"A. Hác","doi":"10.1109/HICSS.1992.183149","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183149","url":null,"abstract":"Studies the sensitivity of asynchronous algorithms in disk buffer cache memory. The algorithms allow for writing the files into the buffer cache by the processes. The number of active processes in the system and the length of the queue to the disk buffer cache are considered in sensitivity study. This information is obtained dynamically during the execution of the algorithms. The time for a block to become available for updating in the buffer cache is given. The number of blocks available for updating in the buffer cache is derived. The performance of the algorithms is compared. It is shown that the proposed algorithms allow for better performance than an algorithm that does not use information about the system load.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"23 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121013858","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183139
Ki-Chang Kim, A. Nicolau
Introduces a technique to parallelize nested loops at the fine grain level. It is a generalization of perfect pipelining which was developed to parallelize a single-nested loop at the fine grain level. The authors explain their method, contrast it with other techniques, show the effect of their technique on some real examples, and briefly discuss an architecture best suited for exploiting this parallelism.<>
{"title":"N-dimensional perfect pipelining","authors":"Ki-Chang Kim, A. Nicolau","doi":"10.1109/HICSS.1992.183139","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183139","url":null,"abstract":"Introduces a technique to parallelize nested loops at the fine grain level. It is a generalization of perfect pipelining which was developed to parallelize a single-nested loop at the fine grain level. The authors explain their method, contrast it with other techniques, show the effect of their technique on some real examples, and briefly discuss an architecture best suited for exploiting this parallelism.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121240822","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183441
A. N. Refenes, Magali E. Azema-Barac, S. A. Karoussos
The paper describes a neural network system for forecasting time series and its application to a non-trivial task in forecasting currency exchange rates. The architecture consists of a two-layer backpropagation network with a fixed number of inputs modelling a window moving along the time series in fixed steps to capture the regularities in the underlying data. Several network configurations are described and the results are analysed. The effect of varying the window and step size is also discussed as are the effects of overtraining. The error backpropagation network was trained with currency exchange data for the period 1988-9 on hourly updates. The first 200 trading days were used as the training set and the following three months as the test set. The network is evaluated both for long term forecasting without feedback (i.e. only the forecast prices are used for the remaining trading days) and for short term forecasting with hourly feedback. By careful network design and analysis of the training set, the backpropagation learning procedure is an active way of forecasting time series. The network learns the training set near perfect and shows accurate prediction, making at least 20% profit on the last 60 trading days of 1989.<>
{"title":"Currency exchange rate forecasting by error backpropagation","authors":"A. N. Refenes, Magali E. Azema-Barac, S. A. Karoussos","doi":"10.1109/HICSS.1992.183441","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183441","url":null,"abstract":"The paper describes a neural network system for forecasting time series and its application to a non-trivial task in forecasting currency exchange rates. The architecture consists of a two-layer backpropagation network with a fixed number of inputs modelling a window moving along the time series in fixed steps to capture the regularities in the underlying data. Several network configurations are described and the results are analysed. The effect of varying the window and step size is also discussed as are the effects of overtraining. The error backpropagation network was trained with currency exchange data for the period 1988-9 on hourly updates. The first 200 trading days were used as the training set and the following three months as the test set. The network is evaluated both for long term forecasting without feedback (i.e. only the forecast prices are used for the remaining trading days) and for short term forecasting with hourly feedback. By careful network design and analysis of the training set, the backpropagation learning procedure is an active way of forecasting time series. The network learns the training set near perfect and shows accurate prediction, making at least 20% profit on the last 60 trading days of 1989.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115955367","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183191
A. Wilson, M. Teller, T. Probert, Dyung Le, R. LaRowe
The Lynx/Galactica Net system forms a clustered, shared memory multiprocessor with scalable performance. The Lynx component is a single-board multiprocessor using advanced, three-dimensional packaging technology to put four high performance RISC processors, memory, and I/O on a single 6m VME board. Galactica Net consists of a high speed mesh network combined with memory update hardware to maintain cache and memory coherence. Multiple Lynx computer nodes are connected together by Galactica Net to form shared memory systems with thousands of processors. A key feature of the Lynx/Galactica Net system is a tight coupling between the operating system and the coherency protocol for efficient, low-overhead coherency. An indication of the expected performance is provided through simulations using hardware-collected parallel address traces.<>
{"title":"Lynx/Galactica Net: a distributed, cache coherent multiprocessing system","authors":"A. Wilson, M. Teller, T. Probert, Dyung Le, R. LaRowe","doi":"10.1109/HICSS.1992.183191","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183191","url":null,"abstract":"The Lynx/Galactica Net system forms a clustered, shared memory multiprocessor with scalable performance. The Lynx component is a single-board multiprocessor using advanced, three-dimensional packaging technology to put four high performance RISC processors, memory, and I/O on a single 6m VME board. Galactica Net consists of a high speed mesh network combined with memory update hardware to maintain cache and memory coherence. Multiple Lynx computer nodes are connected together by Galactica Net to form shared memory systems with thousands of processors. A key feature of the Lynx/Galactica Net system is a tight coupling between the operating system and the coherency protocol for efficient, low-overhead coherency. An indication of the expected performance is provided through simulations using hardware-collected parallel address traces.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116349547","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183378
Nic Stuchfield, B. Weber
Traditional management accounting systems are often unable to provide profitability information relevant to strategic management decisions. The problem is intensified in unstable environments, where changing margins due to deregulation and new entrants, new products, and customer unbundling of firms' offerings can combine to leave unprepared firms with growing numbers of loss-making client relationships. With information on the profitability of customer relationships, managers can identity and defend their most attractive market segments, and aim to turn loss-making accounts into profitable ones. BZW, a London-based securities house, developed an innovative information system that combines activity-based accounting principles and a model of customer profitability based on an income assignment for each of the 6000 trades the firm makes in a day. The system's strategic value appears to be considerable, and industry profitability data recently published will enable us to test the value hypothesis.<>
{"title":"Modeling the profitability of customer relationships: development and impact of Barclays de Zoete Wedd's BEATRICE","authors":"Nic Stuchfield, B. Weber","doi":"10.1109/HICSS.1992.183378","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183378","url":null,"abstract":"Traditional management accounting systems are often unable to provide profitability information relevant to strategic management decisions. The problem is intensified in unstable environments, where changing margins due to deregulation and new entrants, new products, and customer unbundling of firms' offerings can combine to leave unprepared firms with growing numbers of loss-making client relationships. With information on the profitability of customer relationships, managers can identity and defend their most attractive market segments, and aim to turn loss-making accounts into profitable ones. BZW, a London-based securities house, developed an innovative information system that combines activity-based accounting principles and a model of customer profitability based on an income assignment for each of the 6000 trades the firm makes in a day. The system's strategic value appears to be considerable, and industry profitability data recently published will enable us to test the value hypothesis.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116929093","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183287
E. Tick, D. Park
A software visualization tool is described that transforms program execution trace data from a multiprocessor into a single, color image. The image is essentially the program's logical procedure-invocation tree, displayed radially from the root. An algorithm is described that condenses the image both radially and laterally, producing a color-dense abstraction of the program's execution behavior: a program signature, within a workstation window. An implementation of the tool was made in X-Windows, including a user interface that can zoom the image and annotate any node in the image with the corresponding procedure-invocation data. Experimentation with the system was performed with trace data from Panda, a concurrent logic programming system on shared-memory multiprocessors. The authors demonstrate how the tool helps the programmer develop intuitions about parallel performance and how condensation successfully abstracts very large traces.<>
{"title":"Kaleidoscope visualization of fine-grain parallel programs","authors":"E. Tick, D. Park","doi":"10.1109/HICSS.1992.183287","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183287","url":null,"abstract":"A software visualization tool is described that transforms program execution trace data from a multiprocessor into a single, color image. The image is essentially the program's logical procedure-invocation tree, displayed radially from the root. An algorithm is described that condenses the image both radially and laterally, producing a color-dense abstraction of the program's execution behavior: a program signature, within a workstation window. An implementation of the tool was made in X-Windows, including a user interface that can zoom the image and annotate any node in the image with the corresponding procedure-invocation data. Experimentation with the system was performed with trace data from Panda, a concurrent logic programming system on shared-memory multiprocessors. The authors demonstrate how the tool helps the programmer develop intuitions about parallel performance and how condensation successfully abstracts very large traces.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117137280","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183233
H. Hartig, W.E. Kuhnhauser, W. Lux, W. Reck
While many persistent object systems are built on top of conventional operating systems, the BirliX approach has followed the opposite direction: it has built a Unix compatible operating system by emulating the Unix system interface on top of a persistent object system. The BirliX kernel is basically an abstract data type management system. Its basic services are the definition of abstract data types, their instantiation, their identification, and the communication between instances. All abstract data types share a common set of type-independent attributes and methods inherited from the kernel-defined primary type. These methods have been designed to provide mechanisms that are considered useful to build fault tolerant and secure applications in a distributed environment. The Unix emulation on top of the kernel takes advantage of the underlying persistent abstract data type management system.<>
{"title":"Operating system(s) on top of persistent object systems-the BirliX approach","authors":"H. Hartig, W.E. Kuhnhauser, W. Lux, W. Reck","doi":"10.1109/HICSS.1992.183233","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183233","url":null,"abstract":"While many persistent object systems are built on top of conventional operating systems, the BirliX approach has followed the opposite direction: it has built a Unix compatible operating system by emulating the Unix system interface on top of a persistent object system. The BirliX kernel is basically an abstract data type management system. Its basic services are the definition of abstract data types, their instantiation, their identification, and the communication between instances. All abstract data types share a common set of type-independent attributes and methods inherited from the kernel-defined primary type. These methods have been designed to provide mechanisms that are considered useful to build fault tolerant and secure applications in a distributed environment. The Unix emulation on top of the kernel takes advantage of the underlying persistent abstract data type management system.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115496705","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183251
H. Mills
Software is either correct or incorrect in design to a specification in contrast to hardware which is reliable to a certain level to a correct design. Software of any size or complexity can only be tested partially, and typically a very small fraction of possible inputs are actually tested. Certifying the correctness of such software requires two conditions, namely (1) statistical testing with inputs characteristic of actual usage, and (2) no failures in the testing. If any failures arise in testing or subsequent usage, the software is incorrect, and the certification invalid. If such failures are corrected, the certification process can be restarted, with no use of previous testing.<>
{"title":"Certifying the correctness of software","authors":"H. Mills","doi":"10.1109/HICSS.1992.183251","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183251","url":null,"abstract":"Software is either correct or incorrect in design to a specification in contrast to hardware which is reliable to a certain level to a correct design. Software of any size or complexity can only be tested partially, and typically a very small fraction of possible inputs are actually tested. Certifying the correctness of such software requires two conditions, namely (1) statistical testing with inputs characteristic of actual usage, and (2) no failures in the testing. If any failures arise in testing or subsequent usage, the software is incorrect, and the certification invalid. If such failures are corrected, the certification process can be restarted, with no use of previous testing.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114281254","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 : 1992-01-07DOI: 10.1109/HICSS.1992.183308
C. Quinn
The problem of determining the appropriate action in using a new interface, or a new feature of a familiar interface, can be characterized as a form of problem-solving. Using distributed cognition as leverage to approach this problem, explorability is proposed as a framework within which to examine the inferences required to map the desired goal onto an appropriate action. Explorability is supported by consistency. affordances, and culture, each a constraint limiting the space of possible inferences. The three components are described, and the interaction of these three concepts is examined both as a problem and an opportunity for interface designers. The explorability framework is examined as an analytical tool to make explanations and predictions for interface designs.<>
{"title":"Explorability: inferences at the interface","authors":"C. Quinn","doi":"10.1109/HICSS.1992.183308","DOIUrl":"https://doi.org/10.1109/HICSS.1992.183308","url":null,"abstract":"The problem of determining the appropriate action in using a new interface, or a new feature of a familiar interface, can be characterized as a form of problem-solving. Using distributed cognition as leverage to approach this problem, explorability is proposed as a framework within which to examine the inferences required to map the desired goal onto an appropriate action. Explorability is supported by consistency. affordances, and culture, each a constraint limiting the space of possible inferences. The three components are described, and the interaction of these three concepts is examined both as a problem and an opportunity for interface designers. The explorability framework is examined as an analytical tool to make explanations and predictions for interface designs.<<ETX>>","PeriodicalId":103288,"journal":{"name":"Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114592577","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：