[77] | Solomon Atnafu, Lionel Brunie, Harald Kosch, Similarity-Based Operators in Image Database Systems, In Advances in Web-Age Information Management. Second International Conference, WAIM 2001, Xi'an, China, July 9-11, 2001. Proceedings (XS Wang, G Yu, H Lu, eds.), Springer Verlag, Berlin [u. a.], pp. 14-25, 2001.
[bib] |
[76] | Christian Weiß, Hermann Hellwagner, Linda Stals, Ulrich Rüde, Data Locality Optimizations to Improve The Efficiency of Multigrid Methods, In Concepts of Numerical Software, NA, NA, pp. 1-10, 2000.
[bib] [pdf] [abstract]
Abstract: Current superscalar microprocessors are able to operate at a peak performance of up to 1 GFlop/sec. However, current main memory technology does not provide the data needed fast enough to keep the CPU busy. To minimize idle times of the CPU, caches are used to speed up accesses to frequently used data. To exploit caches, the software must be aware of them and reuse data in the cache before it is being replaced. Unfortunately, all conventional multigrid codes are not cache-aware and hence exploit less than 10 percent of the peak performance of cache based machines. Our studies with linear PDEs with constant coefficients show that it is possible to speed up the execution of our multigrid method by a large factor and hence solve a Poisson’s equation with one million unknowns in less than 3 seconds. The optimized reuse of data in the cache allows us to exploit 30 percent of the peak performance of the CPU, in contrast to mgd9v for instance, which achieves less than 5 percent on the same machine. To achieve this, we used several techniques like loop unrolling and loop fusion to better exploit the memory hierarchy and the superscalar CPU. We study the effects of these techniques on the runtime performance in detail. We also study several tools which guide the optimizations and help to restructure the code.
|
[75] | Roland Tusch, Harald Kosch, Laszlo Böszörmenyi, VIDEX: Integrated Generic Video Indexing Approach, In Proceedings ACM Multimedia 2000 : Los Angeles, California, October 30 - November 4, 2000 (Shahram Ghandeharizadeh, Shih-Fu Chang, Stephen Fischer, Joseph A Konstan, Klara Nahrstedt, eds.), ACM, Reading, Mass., pp. 448-451, 2000.
[bib] |
[74] | Markus Schordan, Wolfram Amme, Virtual Method Resolution with Typed Alias Graphs, In Proceedings of the 8th International Workshop on Compilers for Parallel Computers (Samuel Midkiff, José Moreira, Manish Gupta, Siddhartha Chatterjee, Jeanne Ferrante, Jan Prins, William Pugh, Chau-Wen Tseng, eds.), Springer, Heidelberg, pp. 151-162, 2000.
[bib] [pdf] |
[73] | Stefan Podlipnig, Laszlo Böszörmenyi, Quality-aware proxy caching for Web videos, Chapter in Distributed and parallel systems (Peter Kacsuk, Gabriele Kotsis, eds.), Kluwer Academic Publishers, Norwell, MA, USA, pp. 195-204, 2000.
[bib][url] |
[72] | Ahmed Moustefaoui, Harald Kosch, Laszlo Böszörmenyi, Lionel Brunie, Multi-Clip Query Optimization in Video Databases, In Proceedings of IEEE Conference Multimedia and Expo. ICME 2000: 30 July - 2 August 2000, New York, USA/ IEEE Signal Processing Society (Sankar Basu, ed.), IEEE Servoce Center, Piscataway, NJ, pp. 363-366, 2000.
[bib][url] [doi] [pdf] [abstract]
Abstract: A multi-clip query requests multiple video clips. We address the multi-clip query optimization problem. We propose a new heuristic called restricted search interval that maximizes clip sharing between queries and consequently reduces the workload of the video server. The experimental results show that the suggested heuristics reduce the server workload by about 68.7% in comparison to a classical heuristic approach.
|
[71] | Harald Kosch, Roland Tusch, Laszlo Böszörmenyi, Alexander Bachlechner, Bernhard Dörflinger, Christian Hofbauer, Carmen Riedler, The SMOOTH Video DB -Demonstration of an Integrated Generic Indexing Approach, In Proceedings ACM Multimedia 2000 : Los Angeles, California, October 30 - November 4, 2000 (Shahram Ghandeharizadeh, Shih-Fu Chang, Stephen Fischer, Joseph Konstan, Klara Nahrstedt, eds.), Addison-Wesley [u. a.], Reading, Mass., pp. 448-451, 2000.
[bib] [pdf] |
[70] | Harald Kosch, Kurt Breidler, Laszlo Böszörmenyi, The Parallel Video Server SESAME-KB, In Distributed and Parallel Systems - From Concepts to Architectures (Peter Kacsuk, Gabriele Kotsis, eds.), Kluwer Academic Publ, Boston, Mass., pp. 151-154, 2000.
[bib] [abstract]
Abstract: This paper presents the parallel video server SESAME-KB, by the means of its architectural design, striping, scheduling, caching and admission policy. Furthermore we compare our prototype to approaches which influenced our realization. Keywords: Parallel Video Server, Parallel Systems and Communication.
|
[69] | Harald Kosch, CORBA, Web and Databases, In Dexa 2000 Workshop Proceedings, W 10, Greenwich (UK), Sept 2000, Tutorial Abstract (A M Tjoa, R Wagner, A Al-Zobaidie, eds.), IEEE CS Press, Greenwich, London, pp. 674, 2000.
[bib] |
[68] | Harald Kosch, Kurt Breidler, Laszlo Böszörmenyi, A Comparative Study of Selected Parallel Video Servers, In 11th International Workshop on Database and Expert Systems Applications (DEXA'00)6-8 September 2000 , Greenwich, London (A M Tjoa, R R Wagner, A Al-Zobaidie, eds.), IEEE CS Press, U.K., Greenwich, pp. 669-673, 2000.
[bib] [doi] [abstract]
Abstract: This paper compares different selected and representative parallel video servers, including our server SESAME-KB, by the means of their architectural design, striping, scheduling, caching and admission policy. It is shown that these parallel video servers can support a high number of concurrent accesses, but will also reveal missing components, as bottlenecks. Finally we point to further developments
|
[67] | Harald Kosch, Renata Slota, Darin Nikolow, Stefan Podlipnig, Kurt Breidler, MMSRS-Multimedia Storage and Retrieval System for a Distributed System, In High-Performance Computing and Networking, 8th International Conference, HPCN Europe 2000 Amsterdam, May 8-10, 2000 Proceedings (Marian Bubak, ed.), Springer Verlag, Berlin [u. a.], pp. 517-524, 2000.
[bib][url] [doi] [abstract]
Abstract: The Multimedia Storage and Retrieval System described in this paper is aimed at storing and retrieving medical multimedia data such as images and videos. There are three significant problems that automatically need to be addressed: providing enough capacity to fit Terabytes of MPEG files, efficient extraction of video fragments from MPEG files and quality of service issues. We are working on two different approaches to build MMSRS. The first one, described in this article, utilizes commercial HSM software for managing the tertiary storage hardware. The second approach that we have started to investigate is an attempt to build a specialised storage management system from scratch.
|
[66] | Harald Kosch, Managing the operator ordering problem in parallel data-Bases, In Future Generation Computer Systems, Elsevier, vol. Vol 16, no. No 3, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom, pp. 665-676, 2000.
[bib] [pdf] [abstract]
Abstract: This paper focuses on parallel query optimization. We consider the operator problem and introduce a new class of execution strategies called Linear-oriented Bushy Trees (LBT). Compared to the related approach of the General Bushy Trees (GBT) a significant complexity reduction of the operator ordering problem can be derived theoretically and demonstrated experimentally (e.g. compared with GBTs, LBTs authorize optimization time improvement that can reach up to 49%) without losing quality. Finally we demonstrate that existing commercial parallel query optimizers need little extension mod ifications in order to handle LBTs.
|
[65] | Günter Hölzl, Laszlo Böszörmenyi, Distributed Federative QoS Resource Managment-Extended version, In Future Generation Computer Systems, Elsevier, vol. Vol. 16, no. Issue 6, No 16, Oxford, United Kingdom, pp. 717-725, 2000.
[bib] [pdf] |
[64] | Hermann Hellwagner, Klaus Leopold, Ralf Schlatterbeck, Carsten Weich, Performance Tuning of Parallel Real-Time Voice Communication Software, In Proceedings Distributed and Parallel Systems (Peter Kascuk, Gabriele Kotsis, eds.), Kluwer Academic Publishers, Norwell, MA, USA, pp. 57-60, 2000.
[bib] [pdf] [abstract]
Abstract: This paper describes an unconventional way to apply a performance analysis tool for parallel programs (Vampir) to understand and tune the performance of the real-time voice and data communication software running on top of Frequentis’ V4 switch. The execution schedule of the strictly time-triggered V4 switching software is computed off-line; analyzing the schedule to identify e.g. performance bottlenecks used to be a complex and time-consuming process. We present our approach to transform the V4 software schedule’s information into Vampir trace files and use this tool’s facilities to provide a visualization of the schedule. A case study illustrates the benefits of this approach.
|
[63] | Hermann Hellwagner, Markus Lachowitz, Matthias Ohlenroth, Exploring the Performance of VI Architecture Communication Features in the Giganet Cluster LAN, In Proceedings International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA´2000), June 26, 2000, Las Vegas, Vol. 5 (Hamid R Arabnia, ed.), CSREA Press, [Athens, Ga.], pp. 2615-2621, 2000.
[bib] [doi] [pdf] [abstract]
Abstract: The Virtual Interface (VI) Architecture standard was developed to satisfy the need for a high-throughput, low-latency communication system required for cluster computing. This paper presents the results of a performance study of one VI Architecture hardware implementation, the Giganet cLAN (Cluster LAN). The focus of the study is to assess and compare the performance of different VI Architecture data transfer modes and specific features that are available to higher-level communication software like MPI, in order to aid the implementor to decide which VI Architecture options to employ for various communication scenarios. Examples of such options include the use of send/receive vs. RDMA data transfers, polling vs. blocking to check completion of communication operations, multiple VIs, completion queues, and scatter capabilities of VI Architecture.
|
[62] | Laszlo Böszörmenyi, Jürg Gutknecht, Gustav Pomberger, eds., The School of Niklaus Wirth, dpunkt.verlag - Copublication with Morgan-Kaufmann, -, pp. 268, 2000.
[bib][url] [abstract]
Abstract: Niklaus Wirth is one of the great pioneers of computer technology and winner of the ACM âs A.M.Turing Award,the most prestigious award in computer science. He has made substantial contributions to the development of programming languages, compiler construction, programming methodology, and hardware design. While working at ETH Zürich, he developed the languages Pascal and Modula-2. He also designed an early high performance workstation, the Personal Computer Lilith, and most recently the language and operating system Oberon. While Wirth has often been praised for his excellent work as a language designer and engineer, he is also an outstanding educator â something for which he is not as well known. This book brings together prominent computer scientists to describe Wirthâs contributions to education. With the exception of some of his colleagues such as Professors Dijkstra, Hoare, and Rechenberg, all of the contributors to this book are students of Wirth. The essays provide a wide range of contemporary views on modern programming practice and also illuminate the one persistent and pervasive quality found in all his work: his unequivocal demand for simple solutions. The authors and editors hope to pass on their enthusiasm for simple engineering solutions along with their feeling for a man to whom they are all so indebted.
|
[61] | Laszlo Böszörmenyi, Computer im Kindergarten?, In APL-Journal, APL-Journal, vol. Jg. 19, no. Nr. 1, Mannheim, Berlin, Nürnberg, Germany, pp. 29-33, 2000.
[bib][url] [pdf] [abstract]
Abstract: Es läuft eine spannende Diskussion darüber, wie früh man Kinder an den Computer heranführen sollte. Jenseits solcher Intellektuellen-Diskussionen machen manche Firmen schon seit Jahren enorme Geschäfte mit Software und Hardware für Kinder im Kindergartenalter. Auch der erste Computer für Babys hat schon seinen Siegeszug auf dem amerikanischen Markt angetreten. Mit den Geschäftemachern zu diskutieren, ist wahrscheinlich nicht viel klüger als gegen Windmühlen zu kämpfen. Diese Firmen könnten ihre Geschäfte nicht machen, wenn es nicht viele Eltern, Großeltern, Pädagogen und Psychologen gäbe, die den Computer kleinen Kindern mit guter Absicht in die Hand geben. Die folgende Diskussion spricht die wohlwollenden Erwachsenen an, die sich für Kinder verantwortlich fühlen und sich nicht entscheiden können, wann und wie Kinder mit dem Phänomen Computer konfrontiert werden sollten.
|
[60] | Harald Kosch, R Slota, Laszlo Böszörmenyi, J Kitowski, J Otfinowski, P Wójcik, A Distributed Medical Information System for Multimedia Data - The first Years Experience of the PARMED Project, In High-Performance Computing and Networking (M Bubak, R Williams, H Afsarmanesh, B Hertzberger, eds.), Springer Verlag, Berlin [u. a.], pp. 543-546, 1999.
[bib] |
[59] | Hermann Hellwagner, Ivan Zoraja, Vaidy Sunderam, SCIPVM: Parallel Distributed Computing on SCI Workstation Clusters, In Concurrency: Practice and Experience, N, A, vol. Vol 11, no. No 3, N, A, pp. 121-138, 1999.
[bib] [pdf] [abstract]
Abstract: Workstation and PC clusters interconnected by SCI (Scalable Coherent Interface) are very promising technologies for high performance cluster computing. Using commercial SBus to SCI interface cards and early system software and drivers, a two-workstation cluster has been constructed for initial testing and evaluation. The PVM system has been adapted to operate on this cluster using raw device access to the SCI interconnect, and preliminary communications performance tests have been carried out. Our preliminary results indicate that communications throughput in the range of 3.5 MBytes/s, and latencies Research supported by the Applied Mathematical Sciences program, Office of Basic Energy Sciences, U. S. Department of Energy, under Grant No. DE-FG05-91ER25105, the National Science Foundation, under Award Nos. ASC-9527186 and ASC-9214149, and the German Science Foundation SFB342. of 620 ¯s can be achieved on SCI clusters. These figures are significantly better (by a factor of 3 to 4) ...
|
[58] | Hermann Hellwagner, Alexander Reinefeld, SCI: Scalable Coherent Interface, Springer Verlag, Berlin [u a ], pp. 490, 1999.
[bib] [abstract]
Abstract: Architecture and Software for High-Performance Compute Clusters.
|
[57] | Laszlo Böszörmenyi, Computer az óvodában?, In Tani-Tani, Tani-Tani, Budapest, Hungary, pp. 4-14, 1999.
[bib] |
[56] | Laszlo Böszörmenyi, Parallel Programming, In APL-Journal, APL-Journal, vol. Jg. 18, no. Nr. 2, Mannheim, Berlin, Nürnberg, Germany, pp. 85-89, 1999.
[bib] |
[55] | Laszlo Böszörmenyi, Introducing OO with Bottom-up Approach, Structured First, In Proceedings of the Workshop on Object-Oriented Technology (Ana Moreira, Serge Demeyer, eds.), Springer Verlag, Berlin, Heidelberg, New York, pp. 340-342, 1999.
[bib] [abstract]
Abstract: Object-Oriented Technology.
|
[54] | Laszlo Böszörmenyi, Robert Eisner, Herbert Groiss, Adding Distribution to a Workflow Managment System, In 10th DEXA 1999 Workshop Proceedings, Sept 1-3 1999, Florence, Italy (Trevor Bench-Capon, Giovanni Soda, A Min Tjoa, eds.), IEEE CS Press, Los Alamitos, Calif. [u. a.], pp. 17-21, 1999.
[bib] [doi] [abstract]
Abstract: The Workflow Management System AltaVista Works has been extended to be able to process workflows in a distributed, partly even parallel manner. Build-time data are replicated over all participating workflow servers. The distribution of workflow execution is based on migration. Instead of the usual static re-partitioning, a highly dynamic management of distribution is applied.
|
[53] | Laszlo Böszörmenyi, Harald Kosch, Markus Schordan, JavaSet - extending Java by persistent sets, In Proceedings of the 3rd International Austrian-Israeli Technion Symposium (Helmut Gutmann, ed.), Austrian Technion Society, Wien, pp. 58-66, 1999.
[bib] [pdf] |