本文選題：多序列關(guān)聯(lián)性分析 + 多序列比對(duì)��； 參考：《華中科技大學(xué)》2013年碩士論文

【摘要】：多序列關(guān)聯(lián)性分析方法是基于多序列比對(duì)思想，分析序列間遠(yuǎn)近關(guān)系及探索序列關(guān)聯(lián)路線的一種策略。隨著序列數(shù)目的不斷增加，現(xiàn)有基于CPU的多序列關(guān)聯(lián)性分析方法已無法滿足實(shí)際應(yīng)用的需求。隨著圖形處理器(GPU)計(jì)算能力的飛速提高，GPU以其流水線工作模式和強(qiáng)大的并行計(jì)算能力，被廣泛應(yīng)用于解決計(jì)算密集型問題，包括提高多序列關(guān)聯(lián)性分析方法的效率。 結(jié)合GPU強(qiáng)大的并行計(jì)算能力，提出并實(shí)現(xiàn)基于GPU的多序列關(guān)聯(lián)性分析方法，從三個(gè)不同角度進(jìn)行并行優(yōu)化。其中對(duì)關(guān)聯(lián)性分析的算法進(jìn)行改進(jìn)，通過對(duì)算法執(zhí)行過程的調(diào)整，解決算法內(nèi)部的數(shù)據(jù)依賴問題；為降低I/O負(fù)載及實(shí)現(xiàn)異步處理，提出基于GPU的數(shù)據(jù)流并行優(yōu)化策略，對(duì)輸入距離矩陣進(jìn)行數(shù)據(jù)分割，并結(jié)合異步處理模式，實(shí)現(xiàn)CPU與GPU的協(xié)同并行處理；基于GPU的指令流優(yōu)化策略實(shí)現(xiàn)對(duì)不同線程粒度的動(dòng)態(tài)調(diào)用，解決在未知多序列關(guān)聯(lián)關(guān)系的情況下，線程擁塞和線程空載等問題。同時(shí)設(shè)計(jì)基于并行雙調(diào)排序的最小鏈模型，通過并行遍歷子矯正距離矩陣，將遍歷結(jié)果存入最小鏈數(shù)組以進(jìn)行雙調(diào)排序，快速定位當(dāng)前狀態(tài)下的最小值結(jié)點(diǎn)對(duì)，對(duì)多序列關(guān)聯(lián)性分析方法中最耗時(shí)的處理過程進(jìn)行了并行優(yōu)化。 基于Linux操作系統(tǒng)和CUDA平臺(tái)，采用C、C++等語言，實(shí)現(xiàn)基于GPU的多序列關(guān)聯(lián)性分析方法。在保證輸出結(jié)果精確度不變的情況下，減少了輸入數(shù)據(jù)的I/O傳輸時(shí)間，降低了尋找最小值結(jié)點(diǎn)對(duì)的時(shí)間開銷，實(shí)驗(yàn)整體性能與基于CPU的多序列關(guān)聯(lián)分析方法相比，加速比達(dá)到25.1，且具有更穩(wěn)定、更快速的關(guān)聯(lián)性分析性能。
[Abstract]:Multi-sequence correlation analysis is a strategy based on the idea of multi-sequence alignment to analyze the distance and near relationship between sequences and to explore the route of sequence association. With the increasing number of sequences, the existing multi-sequence correlation analysis methods based on CPU can not meet the needs of practical applications. With the rapid improvement of GPU computing power, GPU is widely used to solve computationally intensive problems, including improving the efficiency of multi-sequence correlation analysis with its pipelined mode and powerful parallel computing capability. Combined with the powerful parallel computing ability of GPU, a multi-sequence correlation analysis method based on GPU is proposed and implemented, and parallel optimization is carried out from three different angles. In order to reduce the I / O load and realize asynchronous processing, the parallel optimization strategy of data flow based on GPU is proposed. The input distance matrix is partitioned and the asynchronous processing mode is combined to realize the collaborative parallel processing between CPU and GPU. The instruction flow optimization strategy based on GPU realizes the dynamic call to different thread granularity. In the case of unknown multi-sequence association, the problem of thread congestion and thread no-load is solved. At the same time, the minimum chain model based on parallel bimodal sorting is designed. The traversal result is stored in the minimum chain array to sort the minimum value node pairs in the current state by parallel traversal subcorrecting distance matrix. Parallel optimization of the most time-consuming processing process in the multi-sequence correlation analysis method is carried out. Based on Linux operating system and CUDA platform, the method of multi-sequence correlation analysis based on GPU is realized by using C + C and other languages. The I / O transmission time of the input data is reduced and the time cost of finding the minimum node pair is reduced. The overall performance of the experiment is compared with that of the multi-sequence association analysis method based on CPU. The speedup ratio is 25. 1, and it has more stable and fast correlation analysis performance.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TP332

【參考文獻(xiàn)】

相關(guān)期刊論文 前2條

1 馬海晨;韋剛;吳百峰;;基于GPGPU的生物序列快速比對(duì)[J];計(jì)算機(jī)工程;2012年04期

2 林江;唐敏;童若鋒;;GPU加速的生物序列比對(duì)[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2010年03期

，

本文編號(hào)：1962589