發(fā)布時間：2018-06-27 00:28

  本文選題：大數(shù)據(jù) + 數(shù)據(jù)遷移��； 參考：《哈爾濱理工大學學報》2015年04期

【摘要】：針對目前國內(nèi)銀行的OLAP(Online Analytical Processing)技術滯后于海量數(shù)據(jù)處理需求的情況,研究了海量數(shù)據(jù)存儲、處理的關鍵技術,設計并實現(xiàn)了用Hadoop平臺和關系型數(shù)據(jù)庫架構的銀行海量數(shù)據(jù)處理系統(tǒng).介紹了系統(tǒng)架構、文件存儲層次設計、數(shù)據(jù)遷移方案、任務調(diào)度算法等關鍵技術的實現(xiàn).通過實驗驗證,系統(tǒng)能夠?qū)崿F(xiàn)數(shù)據(jù)的高可靠性與數(shù)據(jù)的高可用性目標,既能滿足業(yè)務數(shù)據(jù)實時處理的需求,也能滿足海量數(shù)據(jù)統(tǒng)計分析的需求.
[Abstract]:In view of the fact that the OLAP (online Analytical processing) technology of domestic banks lags behind the demand for massive data processing, the key technologies of mass data storage and processing are studied. A bank mass data processing system based on Hadoop platform and relational database architecture is designed and implemented. The system architecture, file storage hierarchy design, data migration scheme, task scheduling algorithm and other key technologies are introduced. The experimental results show that the system can achieve the goal of high reliability and high availability of data, which can not only meet the requirements of real-time processing of business data, but also meet the needs of statistical analysis of massive data.
【作者單位】： 哈爾濱金融學院網(wǎng)絡信息中心;
【基金】：黑龍江省科技攻關項目(GC12A307)
【分類號】：TP311.13;TP333

【參考文獻】

相關期刊論文 前10條

1 董延華,李欣,董靜薇,王慕坤;用VB實現(xiàn)大數(shù)據(jù)塊CRC校驗碼算法[J];哈爾濱理工大學學報;2002年01期

2 張裔智;趙毅;湯小斌;;MD5算法研究[J];計算機科學;2008年07期

3 陳全;鄧倩妮;;異構環(huán)境下自適應的Map-Reduce調(diào)度[J];計算機工程與科學;2009年S1期

4 王珊;王會舉;覃雄派;周p，

本文編號：2071985