Refine
Year of publication
Document Type
- Article (2171)
- ZIB-Report (1758)
- In Proceedings (1320)
- Master's Thesis (284)
- Book chapter (166)
- Doctoral Thesis (163)
- Other (99)
- Bachelor's Thesis (69)
- In Collection (66)
- Book (65)
Keywords
- integer programming (36)
- mixed integer programming (34)
- KOBV (33)
- optimal control (32)
- Annual Report (28)
- Kooperativer Bibliotheksverbund Berlin-Brandenburg (24)
- Mixed Integer Programming (21)
- Bibliotheksverbund (20)
- Integer Programming (19)
- mixed-integer programming (19)
Institute
- Numerical Mathematics (1493)
- Mathematical Optimization (1271)
- Visual and Data-centric Computing (1138)
- Visual Data Analysis (1017)
- ZIB Allgemein (951)
- Computational Nano Optics (496)
- Distributed Algorithms and Supercomputing (489)
- Modeling and Simulation of Complex Processes (389)
- Visual Data Analysis in Science and Engineering (378)
- Therapy Planning (315)
Requirements and Design Concepts. Common Graphics Manager (CGM) is the name given to the implementation of GKS, Level 2b, developed at the Free University of Berlin (1,6). This paper is a survey over the "early GKS implementation phase" 1982. Work commenced in February 1980. At the outset some basic design decisions were necessary on account of the special scientific computer environment in Berlin, because the "Berlin GKS" was intended to be a common graphical software package for all the machines.
The paper presents a new uniqueness theory for ODE initial value problems, derived in view of numerical stiff integration. The theory supplies stepsize bounds for stiff integrators that can easily be estimated in extrapolation methods. The additional devices lead to a significant speed-up of computations - in particular in combustion PDE problems.
Tätigkeitsbericht 1986
(1987)
We derive and analyze the hierarchical basis-multigrid method for solving discretizations of self-adjoint, elliptic boundary value problems using piecewise linear triangular finite elements. The method is analyzed as a block symmetric Gauß- Seidel iteration with inner iterations, but it is strongly related to 2-level methods, to the standard multigrid V-cycle, and to earlier Jacobi-like hierarchical basis methods. The method is very robust, and has a nearly optimal convergence rate and work estimate. It is especially well suited to difficult problems with rough solutions, discretized using highly nonuniform, adaptively refined meshes.
Das vorliegende Handbuch wendet sich an alle Benutzer des für den Norddeutschen Vektorrechnerverbund im ZIB installierten Rechners CRAY X-MP/24. Ein großer Teil der Darstellung ist der Optimierung von Programmen gewidmet, die in FORTRAN geschrieben sind. Gerade weil die CRAY einer der schnellsten Rechner der Welt ist, wird jeder Benutzer früher oder später selbst auf dieses Thema stoßen, denn die Maschine ist vornehmlich für Probleme bestimmt, die für die meisten anderen Rechner zu komplex sind. Maschinen der CRAY-Leistungsklasse beziehen einen erheblichen Teil ihrer Geschwindigkeit aus der parallelen Arbeit vieler Komponenten. Programme, die die jeweilige Architektur der Maschine berücksichtigen, können um Größenordnungen schneller (also wirtschaftlicher) sein als Programme, deren Autor beim Schreiben einen sogenannten "klassischen" Rechner vor seinem geistigen Auge hatte. Daher wird auch der technische Aufbau der CRAY X-MP im Handbuch erläutert. Weitere Teile des Handbuches beschäftigen sich mit den Vorrechnern und dem Rechnernetz.
Portable Standard LISP (PSL), a dialect of LISP developed at the University of Utah, has been implemented and optimized for the CRAY 1 and CRAY X-MP supercomputers. This version uses a new implementation technique that permits a step-by-step development of the PSL kernel. The initial CRAY version was acceptable, although the execution speed of the PSL was not as fast as had been anticipated. CRAY-specific optimizations were undertaken that in some cases provided a ten-fold speed improvement, resulting in a fast LISP implementation.
Portable Standard LISP (PSL) is a portable implementation of the programming language LISP constructed at the University of Utah. The version 3.4 of PSL was implemented for CRAY X-MP computers by Konrad-Zuse-Zentrum Berlin; this implementation is based to an important part on the earlier implementation of PSL 3.2 at the University of Utah, Los Alamos National Laboratories and CRAY Research Inc. at Mendota Heights. During the work on implementing PSL the language LISP was investigated for areas which can be supported by vector hardware. One area was found in the COMMON LISP sequence functions and some typical application areas of LISP programming can be improved by vector processing too. A model for the implementation of vector instructions in LISP was developed. For arithmetic an experimental vectorizing extent of the PSL compiler was constructed. With this means full vector hardware capacity can become available for LISP applications.
Tätigkeitsbericht-1985.
(1986)
New, highly efficient techniques for the numerical simulation and parameter identification of large chemical reaction systems are surveyed. The survey particularly addresses to chemists, which are understood to be potential users of the distributed software packages. These packages are written in the form of interactive dialogue systems - thus enabling chemists to concentrate on the chemistry of their problem. Large scale examples from chemical research environments are included.
Tätigkeitsbericht 1985
(1986)