高度并行運算 的英文怎麼說

中文拼音 [gāodùbīnghángyùnsuàn]
高度并行運算 英文
highly-parallel arithmetic
  • 高 : Ⅰ形容詞1 （從下向上距離大; 離地面遠） tall; high 2 （在一般標準或平均程度之上; 等級在上的） above...
  • 度 : 度動詞[書面語] （推測; 估計） surmise; estimate
  • 行 : 行Ⅰ名詞1 （行列） line; row 2 （排行） seniority among brothers and sisters:你行幾? 我行三。where...
  • 運 : Ⅰ動詞1 （物體位置不斷變化） move; revolve 2 （搬運; 運輸） carry; transport 3 （運用） use; wield...
  • 算 : Ⅰ動詞1 （計算數目） calculate; reckon; compute; figure 2 （計算進去） include; count 3 （謀劃;計...
  • 高度 : altitudeheightelevation
  • 運算 : [數學] operation; arithmetic; operating
  1. Through the simulation of large - scale circuit simulation proved that use the crossover tearing technology could detailed network structure, simplify the diagnostic process, and the neural network can parallel deal with the diagnosis information, and the logic operation can judge the information of the multi - fault. the illustrative simulation shows that it can increase the diagnosis speed and decrease the workload before test

    通過對大規模模擬電路的模擬證明，使用交叉撕裂明細網路結構，簡化診斷過程，且運用神經網路組對信息進行并行處理，邏輯分析運算對多故障信息進行處理判斷，大大提高了故障診斷速度，減小了測前工作量。
  2. In regard to the knowledge by learning, we present that the super - dimensional cube, which has the characteristic to integrate various knowledge representation conveniently, is used as the architecture of knowledge base. the speediness algorithm of knowledge access, the best matching cube selection algorithm, is discussed in this paper when the super - dimensional cube is used as the architecture of knowledge base. the time complexity of the algorithm is analyzed, and the parallel algorithm is discussed also

    在這篇文章中我們提出一種多智能體系統新的協商模型：基於「這是什麼」 （或「要做什麼」 ）學習機制的協商而不是常規的「怎麼做」模型；學習得到的知識，為了方便集成多種知識表示方法，我們運用高維方體的知識庫體系結構，在此基礎上進行快速訪問演算法的設計，分析了演算法的復雜度，並討論了演算法的并行實現；在模型中，我們還根據功能的需要與環境的需要，提出了智能體的三維模型。
  3. In order to enhance coding / decoding speed, we have rewritten image wavelet decomposition and composition program with c6701 assembly language in which we made use of parallel instructions and pipeline optimizing skill to lessen calculating time

    為了提高編解碼速度，圖像小波分解與合成程序完全用c6701匯編指令編寫，其中大量使用并行指令和流水線優化，使運算時間大大減少。
  4. The digitizer based on pxi bus uses fpga ( field programmable gate array ) to implement 256 points, radix - 2 dit fft ( fast fourier transform algorithm ). the design uses pipelining for fft processing and can accomplish sampling and processing signals of two channels at the same time. in the signal acquisition circuit, - a / d convector is used to enhance the precision of the signal sampling

    在本設計中，採用fpga （ fieldprogrammablegatearray ）實現了256點基2dit演算法復數fft （ fastfouriertransformalgorithm快速傅氏變換演算法）處理器，具有較高的速度和運算精度fft ，設計採用流水線處理方式大大的提高了處理速度，可實現對兩個通道輸入信號的并行採集與處理。
  5. Digital - optical hybrid image processing, optical pattern recognition ( opr ) technique performs image processing and recognition through optical correlation and digital processing. it has promising application prospect in automatic target recognition because of its outstanding merits of high - speed parallel operation, high spatial resolution, flexibility and accuracy

    光電混合圖像處理、光學模式識別（ opr ）技術以光學相關與數字處理相結合的方式實現對圖像信號的處理與識別，因而具有光速運算、高速并行處理、高解析度與計算機的靈活性、精確性的突出優點，在自動目標識別（ atr ）領域有著良好的應用前景。
  6. The experimental results indicate that distributed parallel computing on multiple computers is faster than sequent computing on a single computer for complex problem on computing

    實驗結果充分說明：大計算量的并行性問題用分散式并行計算系統進行計算能夠提高其運行速度。
  7. For example, we modify the search algorithm in the schedule system to improve its run - time performance, and implement a in - kernel data delivering mechanism in the relay system to increase its capability

    這其中主要包括，在調度系統中通過改進節點搜索演算法，提高整個調度過程的運行效率;在中繼系統中通過實現基於內核的數據轉發模式，提高中繼系統的并行處理能力。
  8. Because the performance of single microprocessor can " t be improved indefinitely due to the limitation of speed and machining technology of chips, so came the massive parallel processors system

    而單處理器計算機系統由於處理器運算性能受限於晶元速度極限和加工工藝極限，不可能無限提高。於是超大規模并行處理系統應運而生。
  9. Abstract : an efficient parallel global router using random optimization that is independent of net ordering is roposed. parallel approaches are described and strategies guaranteeing the routing quality are discussed. the wire length model is implemented on multiprocessor, which enables the algorithm to approach feasibility of large - scale problems. timing - driven model on multiprocessor and wire length model on distributed processors are also presented. the parallel algorithm greatly reduces the run - time of routing. the experimental results show good speedups with no degradation of the routing quality

    文摘:設計實現了一個高效的線長模式下基於多處理機的并行總體布線器.通過對非時延驅動模式下串、并行演算法的總運行時間和求解時間的比較，表明該并行演算法能夠在保證求解質量無明顯變化的前提下大大加快總體布線演算法的求解速度.同時，也提出了基於分散式體系結構的并行總體布線演算法
  10. The 3d turbulent flow fields around high - speed trains of four kinds of model uain have been calculated by 3d parallel compute program coded with the algorithms presented in this thesis. from the calculation results, the aerodynamic drag of different types of model trains are obtained, some characteristics of the flow field over the high - speed train are analyzed preliminarily such as pressure distribution and diversification, velocity distribution and diversification, and the affections of different train head shape to pressure. velocity, pressure gradient and the like

    用三維并行計算程序模擬了四種不同高速列車車型的外流場，獲得了不同車型的運行阻力；詳細分析了列車附近流場的壓力分佈及其變化情況，討論了列車頭型對列車通過時產生的壓力波動的影響；分析了四種車型外部流場的速度分佈規律及其變化情況，分析了列車頭部形狀對它們的影響。
  11. 3 ) using the capability of tms320c5409, all the algorithms of separating a target from its background, extracting the character and identifying the target are operated in " parallel ". for example, during the separation of target from its background the characters of the target were picked up

    利用tms320c5409的特點，將目標分離、特徵提取、目標識別等演算法的操作「并行」進行，如在目標和背景分離的過程中提取目標的特徵量，大幅度提高了系統的運算速度。
  12. In view of the commonly - used block cipher and public key cryptography, this article focuses on the improvement we have made for the algorithms about aes, dbs, ecc and rsa. based on the dsp chip tms320c6201 from ti company, we have accomplished the speed of 0. 92 usec per aes encryption / decryption, 1. 335 usec per dbs encryption / decryption, and 3

    針對目前信息安全體系中常用的分組密碼和公鑰密碼演算法，本文著重改進了des 、 aes的快速演算法實現，並在基於ti公司dsp晶元tms320c6201的系統中實現了0 . 92usec加（解）密一個aes分組， 1 . 335usec加（解）密一個des分組， 3 . 585usec加（解）密一個tdea分組；同時通過改造，使montgomery約化更易於進行并行處理，並通過改進橢圓曲線密碼（ ecc ）底層運算和倍點演算法，使得rsa和橢圓曲線密碼的實現速度得到了極大的提高。
  13. After that, according to the technical features of tms320 dm642 dsp, we elaborately design the program flow and data flow of the encoder. we make use of dma to transfer data to alleviate the burden of dsp core. further more we utilize many code optimization techniques to take full advantages of the parallel computation units

    之後通過分析ti公司的tms320dm642dsp的技術特點，對編碼器的程序流程和數據流進行了精心的設計，採用了dma傳輸數據以減輕dsp內核的負擔，而且採用了多種程序代碼級優化方法來充分利用dsp內核的并行計算單元，提高程序的運行速度。
  14. Hunan limited - region numerical weather prediction system is based upon the fifth - generation psu / ncar meso - scale model ( version 3. 4 ), named mm5 which is a non - hydrostatic model. the initial and lateral boundary conditions of the system are generated from the output of the t213 model of the national meteorology center ( nmc ) and the routine observation data of surface and high levels

    湖南省有限區域數值天氣預報系統是以psu / ncar的mm5v3 . 4 （第五代中尺度模式版本3 . 4 ）非靜力中尺度模式為基礎，與國家氣象中心t213模式輸出場資料連接，並應用最臨近常規高空、地面觀測資料做初始場及參數修訂，在銀河機群服務器系統yh / cs16并行運算環境下運行的中尺度數值模式。
  15. The paper makes great efforts on the software optimization of evrc vocoder. based on the understanding of tms320c64xx cpu structure, we do deeply - optimization on the loop which appear usually in voice signal processing, and this improve the utility ratio of cpu and the parallelity degree of cpu function cell. at the same time, we utilize the bit - exact test to test the fixed - point evrc vocoder with the test vectors of tia / eia / is - 718, which improve the robustness of the vocoder

    本文圍繞定點evrc聲碼器的軟體優化，做了大量的工作，在充分理解tms320c64xxcpu結構的基礎上，針對語音信號處理中大量出現的循環運算進行了深度優化，大大提高了cpu的利用率以及cpu功能單元的并行程度，同時，我們還用tia / eia / is - 718的測試向量對定點evrc聲碼器進行了嚴格比特對準測試，提高了聲碼器的魯棒性。
  16. Storing information in molecules of dna could allow for an information density of approximately 1 bit per cubic nanometer. the energy consumed by dna molecular biology computing is billionth of that consumed by one conventional computer. the characteristics of dna molecular biology computing mentioned above which are high parallelism, huge capability and low consumption are incomparable and irreplaceable to the existing computers and parallel ones

    Dna分子生物演算法具有高度的并行性，運算速度快； dna作為信息的載體其貯存的容量非常之大， 1立方米的dna溶液可存儲1萬億億的二進制數據； dna分子生物計算所消耗的能量只有一臺電子計算機完成同樣計算所消耗的能量的十億分之一； dan計算的上述特性，即運算的高度并行性、大容量、低消耗是目前計算機和并行計算機所無法比擬和替代的。
  17. In this method, each routing node runs a same ospf dameon and they do the routing computation and protocol processing independently based on their own warr. ospf damenoas commuciate by message delivery, with little semaphores needed to do the synchronization work between the dameons, and so we can get highly parallel degree between the dameons

    該方案中，各個節點運行相同的ospf進程，並基於各自warr獨立進行路由計算和協議處理，進程間通過消息傳遞的方式進行通信，不需要太多的信號量進行同步，因此有很高的并行度。
  18. However, when facing complex cases, cfd software requires large memory and cpu time. for ordinary user who has no acess to expensive computer resource, efficient programming is of the most impotant

    然而， cfd技術對計算機的要求非常高，一般空調室內，尤其是幾何空間構造復雜的空調室內環境需要用巨型機作并行運算才能得出精度較高的結果。
  19. The algorithm of upgrading of general face recognition system is based on all serial algorithm in a single computer, it is very slow to deal with a large amount of data, and the efficiency is low. so, this article introduces the application of grid computing in face recognition system, upgrades the original serial algorithm of face data into parallel algorithm in the grid platform by using mpi parallel program, realizes the already existing updated algorithm of the recognition of face to be processed in the distributed computers, which has strengthened the systematic ability to deal with a large amount of data, in order to improve systematic performance

    常規人臉識別系統中的更新演算法都是基於單機的串列演算法，在處理大量數據的時候速度慢，效率低，介紹了網格計算在人臉識別系統中的應用，把原來的人臉數據更新串列演算法改為并行演算法並通過編寫mpi并行程序移植到該網格計算平臺中運行，實現了原有人臉識別系統中更新演算法的分散式處理，增強了系統處理大量數據的能力，以達到提高系統性能的目的。
  20. Because the number of data in image disposal processing is very large, and the existence of many convolution operations and matrix multiplication operations in all kinds of algorithms, it is possible that parallel algorithm design and application can be used in image disposal processing

    由於圖像處理過程中的數據量巨大，而且各種演算法中大量卷積運算和矩陣乘法運算的存在，就為圖像處理過程中的并行演算法設計和實現提供了可能。并行演算法的設計是為了提高圖像處理的速度，在有限的空間和時間處理更多的圖像數據。
分享友人
分享到 Line

分享到臉書