We investigate an ellipsoid algorithm for nonlinear programming. After describing the basic steps of the algorithm, we discuss its computer implementation and present a method for measuring computational efficiency. T...
详细信息
We investigate an ellipsoid algorithm for nonlinear programming. After describing the basic steps of the algorithm, we discuss its computer implementation and present a method for measuring computational efficiency. The computational results obtained from experimenting with the algorithm are discussed and the algorithm's performance is compared with that of a widely used commercial code.
An examination is conducted of measures for evaluating the performance of algorithms for single instruction stream-multiple data stream (SIMD) machines. The SIMD mode of parallelism involves the use of a large number...
详细信息
An examination is conducted of measures for evaluating the performance of algorithms for single instruction stream-multiple data stream (SIMD) machines. The SIMD mode of parallelism involves the use of a large number of processors which are synchronized together. Although all processors execute the same instruction at the same time, each processor operates on a different data item. In general, the complexity of parallel algorithms is a function of the machine size, problem size, and type of interconnection network used to provide communications among the processors. Measures that quantify the effect of changing the machine-size/problem-size/network-type relationships are needed. Several such measures are presented and applied to an example SIMD algorithm from the image processing problem domain. Such measures include execution time, speed, parallel efficiency, overhead ratio, processor utilization, redundancy, cost effectiveness, speed-up of the parallel algorithm over the corresponding serial algorithm, and an additive measure called ''price'' which assigns a weighted value to computations and processors. Figures.
This is a comparison of two state-of-the-art large-scale nonlinear optimization systems exhibiting unprecedented problem solution capabilities both in size of problem handled and method of solution. These codes are MI...
详细信息
This is a comparison of two state-of-the-art large-scale nonlinear optimization systems exhibiting unprecedented problem solution capabilities both in size of problem handled and method of solution. These codes are MINOS, developed by B. A. Murtagh and M. A. Saunders, and XS, developed by G. G. Brown and G. W. Graves. The codes are evaluated with respect to their problem solving capabilities and potential for practical applica- tion by analysts. Computational results are presented for thirteen nonlinear and nonlinear mixed integer test problems with from two to 793 variables (12 to 100 integer variables) and one to 401 constraints. Portions of this work were presented at the CORS/ORSA/TIMS joint meeting in Toronto, May 1981.
A typical organization of magnetic bubble memories is the major-minor loop structure, where n (minor) loops are connected via switches to another (major) loop, containing the I/O station. We work on a sorted file F, s...
详细信息
A typical organization of magnetic bubble memories is the major-minor loop structure, where n (minor) loops are connected via switches to another (major) loop, containing the I/O station. We work on a sorted file F, split into subfiles allocated in the minor loops.
暂无评论