![]() ![]() The utility programs for FITS tables are especially rich and powerful, and provide functions for presentation of file contents, extraction of specific rows or columns, appending or merging tables, binning values in a column or selecting subsets of rows based on a boolean expression. The FTOOLS package contains many utility programs which perform modular tasks on any FITS image or table, as well as higher-level analysis programs designed specifically for data from current and past high energy astrophysics missions. With further optimizations (e.g., more careful load balancing), speeds in excess of today's vector processing computers should be possible.įTOOLS, a highly modular collection of utilities for processing and analyzing data in the FITS (Flexible Image Transport System) format, has been developed in support of the HEASARC (High Energy Astrophysics Research Archive Center) at NASA's Goddard Space Flight Center. The cost/performance ratio for SIMD machines like the Maspar MP-1 make them an extremely attractive alternative to either vector processors or large multiple instruction, multiple data (MIMD) type parallel computers. Our code attains speeds comparable to one head of a Cray Y-MP, so single instruction, multiple data (SIMD) type computers can be used for these simulations. We also simulate the formation of structure in an expanding, model universe using 1,048,576 particles. The first is a simulation of the close passage of two gravitationally, interacting, disk galaxies using 65,636 particles. We apply the code to two problems of astrophysical interest. Roughly 55% of the tree search time is spent in communications between processors. During the tree search, node data that is needed from another processor is simply fetched. The tree search and the summation of accelerations also perform well. Each node in the tree is treated as a monopole. An implementation of this tree-building algorithm on a 16k processor Maspar MP-1 performs well and constitutes only a small fraction (approximately 15%) of the entire cycle of finding the accelerations. This data structure is a completely balanced tree (i.e., each particle is paired with exactly one other particle) and maintains good spatial locality. The sorted lists are recursively divided along their x, y and z coordinates. The tree construction uses fast, parallel sorts. We describe a fast tree algorithm for gravitational N-body simulation on SIMD parallel computers. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |