{"title":"Benchmark workload generation and performance characterization of multiprocessors","authors":"A. Nanda, L. Ni","doi":"10.1109/SUPERC.1992.236713","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236713","url":null,"abstract":"A comprehensive benchmark workload generation and performance characterization methodology is described for multiprocessors supporting the shared-variable computational paradigm. The method can be tailored to meet the selective assessment needs of each individual situation. The approach is based on characterizing a unit grain of computation to generate a desired benchmark workload, and using a family of workload emulation kernels to systematically investigate the effect of each parameter in the workload on the multiprocessor performance. The resultant characterization is independent of any particular application or algorithm.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123574634","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}
C. Practico, F. Hanson, Huihuang Xu, D. Jarvis, Michael S. Vetter
{"title":"Visualization for the management of renewable resources in an uncertain environment","authors":"C. Practico, F. Hanson, Huihuang Xu, D. Jarvis, Michael S. Vetter","doi":"10.1109/SUPERC.1992.236677","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236677","url":null,"abstract":"The management of renewable resources such as commercial and recreational fisheries can be difficult due to lack of data, environmental uncertainty, and a multitude of species interactions. It is noted that electronic visualization is essential for interpreting the multidimensional supercomputer results. Furthermore, visualization can show the resource manager how changes in management policy can affect the overall economic performance of the fishery, as well as how sensitive the performance results are to variations in the poorly known data parameters. An implementation of a 'world within a world' vision concept permits visualization of a 3-D solution surface in an inner world, which can be changed along with a corresponding change of the parameters of a 3-D outer world. This implementation allows the resource manager to visualize multidimensional resources along with parameter sensitivity of the optimal value and optimal control. The research discussed pertains to the Lake Michigan fishery.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125335949","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}
A. Gunzinger, U. A. Müller, W. Scott, B. Bäumle, P. Kohler, Johann vonder Mühll, F. Müller-Plathe, W. V. Gunsteren, W. Guggenbühl
{"title":"Achieving super computer performance with a DSP array processor","authors":"A. Gunzinger, U. A. Müller, W. Scott, B. Bäumle, P. Kohler, Johann vonder Mühll, F. Müller-Plathe, W. V. Gunsteren, W. Guggenbühl","doi":"10.1109/SUPERC.1992.236650","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236650","url":null,"abstract":"The authors discuss the MUSIC (multi-signal-processor system with intelligent communication) system, a parallel distributed memory architecture based on digital signal processors (DSPs). A system with 60 processor elements is operational. It has a peak performance of 3.8 GFlops and an electrical power consumption of less than 800 W (including forced air cooling), and it fits into a 19\" rack. Two applications (the backpropagation algorithm for neural network learning and molecular dynamics simulations) run about six times faster than on a CRAY Y-MP and two times faster than on a NEC SX-3. A sustained performance of more than 1 GFlops is reached.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128297717","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}
{"title":"Virtual worlds for visualizing information","authors":"Steven K. Feiner","doi":"10.1109/SUPERC.1992.236639","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236639","url":null,"abstract":"Virtual worlds are computer-generated environments created by coupling 3-D displays and interaction devices to powerful graphics workstations. The author describes work in the design of virtual worlds being carried out by Columbia's Computer Graphics and User Interfaces Group Two of the main themes of this group's research are exploiting true 3-D interaction and display devices to visualize and manipulate rich information spaces, and using artificial-intelligence techniques to automate the generation of effective graphics. The projects discussed address virtual worlds for visualizing multivariate data, hybrid user interfaces that merge 2-D and 3-D displays and interaction devices, and augmented realities in which the surrounding physical world is annotated with knowledge-based 3-D graphics.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128999825","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}
{"title":"Multiparameter continuation methods for tracking desired flow states","authors":"A. Yeckel, L. Scriven","doi":"10.1109/SUPERC.1992.236701","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236701","url":null,"abstract":"Conventional means of continuation and critical point tracking have been adapted for two novel applications: the tracking of onset of separation at a boundary, and of the limiting conditions for pinning of a static contact line at an edge. The method presented applies more generally to tracking flow states under constrained variation of parameters; the only difference from one application to another is the choice of constraints. Because this choice is virtually limitless, it is believed that the method could have wide utility in computational fluid mechanics and other systems modeling, whether for scientific or for technological purposes.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126678004","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}
{"title":"The role of computational clusters","authors":"E. Brooks, T. Nash, Karl-Heinz Winkler","doi":"10.1109/SUPERC.1992.236698","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236698","url":null,"abstract":"A panel session on the role of computational clusters is presented. Attention is given to experiences with cluster computing at Lawrence Livermore National Laboratory, and to experience with cluster computing at Fermilab.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130654078","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}
{"title":"Applications and performance analysis of a compile-time optimization approach for list scheduling algorithms on distributed memory multiprocessors","authors":"Yeh-Ching Chung, S. Ranka","doi":"10.1109/SUPERC.1992.236653","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236653","url":null,"abstract":"The authors discuss applications of BTDH (bottom-up top-down duplication heuristic) to list scheduling algorithms (LSAs). There are two ways to use BTDH for LSAs. BTDH can be used with an LSA to form a new scheduling algorithm (LSA/BTDH), and it can be used as a pure optimization algorithm for an LSA (LSA-BTDH). BTDH has been applied with two well-known LSAs: the highest level first with estimated time (HLFET) and the earlier task first (ETF) heuristics. Simulation results show that, given a directed acyclic growth (DAG), the graph parallelism of the DAG can accurately predict the number of processors to be used such that a good scheduling length and a good resource utilization (or efficiency) can be achieved simultaneously. In terms of speedups, LSA/BTDH >or= LSA-BTDH >or= ETF >or= HLFET. Experimental results of scheduling FFT programs, which are written in a single program multiple data (SPMD) programming approach, on NCUBE-2 are also presented. The results confirm the simulation results and show that the speedups of LSA/BTDH and LSA-BTDH are better than the speedups of LSAs.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131971258","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}
{"title":"Storage systems for national information assets","authors":"R. Coyne, H. Hulen, R. Watson","doi":"10.1109/SUPERC.1992.236641","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236641","url":null,"abstract":"An industry-led collaborative project, called the National Storage Laboratory (NSL), has been organized to investigate technology for storage systems that will be the future repositories for the national information assets. Lawrence Livermore National Laboratory through its National Energy Research Supercomputer Center (NERSC) is the operational site and the provider of applications. It is anticipated that the integrated testbed system will represent a significant advance in the technology for distributed storage systems capable of handling gigabyte class files at gigabit-per-second data rates. The NSL collaboration is undertaking research in four areas: network-attached storage; multiple, dynamic, distributed storage hierarchies; layered access to storage system services; and storage system management. An overview of the prototype storage system is given. Three application domains have been chosen to test and demonstrate the system's effect on scientific productivity; climatic models, magnetic fusion energy models, and digital imaging.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131030657","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}
{"title":"Scalability analysis of partitioning strategies for finite element graphs: a summary of results","authors":"A. Grama, Vipin Kumar","doi":"10.1109/SUPERC.1992.236707","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236707","url":null,"abstract":"The authors present a scalability analysis of three partitioning strategies, namely striped partitioning, binary decomposition, and scattered decomposition. The analysis is performed using the Isoefficiency metric, which helps in predicting the performances of these schemes on a range of processors and architectures. The performance of each of these schemes is related to the various problem characteristics such as mesh geometry and density. Isoefficiencies are presented for hypercube and mesh connected architectures. Theoretical results are verified through simulations.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130945272","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}
{"title":"The distributed virtual windtunnel","authors":"S. Bryson, Michael J. Gerald-Yamasaki","doi":"10.1109/SUPERC.1992.236675","DOIUrl":"https://doi.org/10.1109/SUPERC.1992.236675","url":null,"abstract":"An implementation of a distributed virtual environment for the shared interactive visualization of large unsteady three-dimensional flowfields is described. Computation of the visualizations is performed on a Convex C3240 computer and the visualization data are transferred over a high-speed network to a Silicon Graphics Iris workstation for rendering. A boom-mounted six-degree-of-freedom head-position-sensitive stereo CRT (cathode ray tube) system is used for display. A hand-position-sensitive glove controller is used for controlling various tracers (e.g., 'smoke') for the visualization of the flow. User commands are sent to the Convex, which interprets these commands and computes the corresponding visualization. With this architecture, several users may share and cooperatively control the visualization generated by the Convex. The distributed architecture is also interesting to those using conventional screen and mouse interfaces.<<ETX>>","PeriodicalId":315340,"journal":{"name":"Proceedings Supercomputing '92","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1992-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132897979","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}