In this article , we introduce the design and implementation of key technologies in pems ( parallel embedded multiprocessing system ) , which is the operating system of the satellite - carried parallel computer system 本文以星載并行嵌入式計算機系統(tǒng)為背景,闡述其并行嵌入式操作系統(tǒng)pems ( parallelembeddedmultiprocessingsystem )中關(guān)鍵技術(shù)研究與實現(xiàn)。
This article has six chapters . chapter 1 introduces the development and the classification of the parallel computer , simply discusses the basic theory of parallel computing , and also introduces research meaning of this article 本文總共分為六章,其內(nèi)容如下:第1章,主要介紹并行計算機的發(fā)展及分類,并行計算的基本理論及本論文所做的主要工作。
Parallel computer is an important area of the research of high - performance computer , and the efficient , rapid , large - capacity interconnection network is the base of large - scale parallel computer 并行計算機是高性能計算機領(lǐng)域的重要研究方向,大規(guī)模并行機的關(guān)鍵是需要高效、快速、大容量的互連網(wǎng)絡(luò),所以改善并行計算機性能的關(guān)鍵問題是設(shè)計高性能的互連網(wǎng)絡(luò)。
The performance model has been validated against the implementation results obtained on a dawing tci700 parallel computer and can be readily used to estimate the performance before the optimistic simulation is implemented 在曙光tc1700并行機上的性能實驗表明,我們的解析性性能分析模型得到的結(jié)果與實際實現(xiàn)的結(jié)果是非常接近的,可以用于樂觀仿真實現(xiàn)前的性能預(yù)測。
It is found that when the number of the grids is over 512000 , the efficiency of two parallel computers is more than 67 . 9 % , and over 52 . 8 % by four computers . and the efficiency will improve as the grid number increases 通過測試發(fā)現(xiàn)當網(wǎng)格數(shù)量超過512000時,兩臺計算機并行效率超過67 . 9 ;四臺計算機并行效率超過52 . 8 ,并且隨著網(wǎng)格數(shù)量增加,并行效率能夠得到進一步提高。
The conflict can be resolved by the parallel compute , but two questions exist at the same time . firstly , the related high rank parallel software is so scarce that the parallel computer can only used by the professional programmer 并行機的出現(xiàn)為復(fù)雜結(jié)構(gòu)精細分析提供了硬件基礎(chǔ),但由于高層次并行軟件的缺乏使得并行計算機往往只能由專業(yè)程序員使用,另外,并行機價格昂貴,不易普及。
The arithmetic architecture of parallel pragram is closely related with the architecture of parallel computer system . presently , there are some popular models such as share - memory model , message passing model and data paralleling model 并行計算采用并行程序?qū)崿F(xiàn),并行程序設(shè)計的算法結(jié)構(gòu)與并行計算機系統(tǒng)結(jié)構(gòu)緊密相關(guān),當前流行的算法結(jié)構(gòu)有共享存儲、消息傳遞、數(shù)據(jù)并行等多種模型。
This article details the arrangement relationship of independence , order , deflection order and the waterfall arrangement ways according to path number and running time , with the stress on the " trapezoid " arithmetic of batch process in parallel computer 摘要介紹了含有獨立、順序、有偏序關(guān)系的批作業(yè)的梳理,以及批作業(yè)按作業(yè)序列路徑數(shù)、運行時間長短排序的瀑布式方法,重點講述了并行機批作業(yè)執(zhí)行時的“梯形”算法。
Ga poses implicit parallelism and is suitable for implementation on large scale parallel computers . dividing the whole population into sub - populations and coarse - grained island model of exchanging information among sub - populations are the most direct parallel method Gas具有天然的并行性,非常適合于在大規(guī)模并行計算機上實現(xiàn),把串行g(shù)as中的單一群體分成多個子群體,各子群體之間相互交換信息的粗粒度并行是將gas并行化的最直接方式。
This article introduces the developing history and the current situation of the automatic goal discern , and on the basis of structure of the parallel computer and neural network knowledge , it analyses the hardware and software engineering foundation that realized the real - time goal recognition system 介紹了自動目標識別的發(fā)展歷史和現(xiàn)狀,在并行計算機結(jié)構(gòu)和神經(jīng)網(wǎng)絡(luò)知識的基礎(chǔ)上,分析了實現(xiàn)實時目標識別系統(tǒng)的硬件和軟件技術(shù)基礎(chǔ)。