【摘要】：由于內(nèi)存數(shù)據(jù)庫(kù)將數(shù)據(jù)存儲(chǔ)在內(nèi)存中,相比于傳統(tǒng)磁盤數(shù)據(jù)庫(kù)在性能方面得到了極大的提升,而且隨著內(nèi)存容量的不斷增加和內(nèi)存價(jià)格的不斷下降,使內(nèi)存數(shù)據(jù)庫(kù)逐漸從一個(gè)設(shè)想變?yōu)楝F(xiàn)實(shí),近年來(lái)關(guān)于內(nèi)存數(shù)據(jù)庫(kù)的研究吸引了人們?cè)絹?lái)越多的關(guān)注,各種關(guān)于內(nèi)存數(shù)據(jù)庫(kù)索引、數(shù)據(jù)存儲(chǔ)、并發(fā)控制等方面的研究成果不斷涌現(xiàn),內(nèi)存數(shù)據(jù)庫(kù)的研究已經(jīng)成為數(shù)據(jù)庫(kù)研究的一個(gè)熱點(diǎn)。同時(shí),各大數(shù)據(jù)庫(kù)廠商為了搶占市場(chǎng)都相繼推出了相關(guān)的內(nèi)存數(shù)據(jù)庫(kù)產(chǎn)品,這也加快了內(nèi)存數(shù)據(jù)庫(kù)應(yīng)用的普及。如今,內(nèi)存數(shù)據(jù)庫(kù)已經(jīng)成為解決系統(tǒng)存儲(chǔ)和檢索性能問題的重要解決方案之一。本文以內(nèi)存數(shù)據(jù)庫(kù)作為研究對(duì)象,提出了一套內(nèi)存數(shù)據(jù)庫(kù)索引新算法,并以網(wǎng)絡(luò)安全管理系統(tǒng)作為應(yīng)用場(chǎng)景,將內(nèi)存數(shù)據(jù)庫(kù)引入并進(jìn)行改進(jìn)設(shè)計(jì),顯著地提高了系統(tǒng)的整體性能。具體成果如下:(1)提出一種新型的內(nèi)存數(shù)據(jù)庫(kù)索引算法DLB+樹,通過將B+樹的葉子結(jié)點(diǎn)擴(kuò)展為雙葉子結(jié)點(diǎn),并改進(jìn)了相關(guān)的操作算法,最終得到一種更適用于范圍查詢且有較高空間利用率的索引算法。實(shí)驗(yàn)證明,當(dāng)查詢數(shù)據(jù)達(dá)到一定數(shù)量,其范圍查詢效率是B+樹查詢效率的2倍。(2)將內(nèi)存數(shù)據(jù)庫(kù)實(shí)際應(yīng)用到網(wǎng)絡(luò)安全管理系統(tǒng)中,改善當(dāng)前系統(tǒng)在數(shù)據(jù)庫(kù)存儲(chǔ)和檢索方面存在的功能和性能缺陷,利用內(nèi)存數(shù)據(jù)庫(kù)在存儲(chǔ)上的靈活性和檢索上的高效性,對(duì)系統(tǒng)用戶管理模塊的單點(diǎn)登錄和權(quán)限控制功能以及安全事件管理模塊的事件匹配和實(shí)時(shí)事件監(jiān)視都做了重新設(shè)計(jì)和改進(jìn),還針對(duì)內(nèi)存數(shù)據(jù)庫(kù)造成的數(shù)據(jù)不一致性情況設(shè)計(jì)了相應(yīng)的持久化方案。(3)對(duì)改進(jìn)后的網(wǎng)絡(luò)安全管理系統(tǒng)做了功能和性能測(cè)試,功能測(cè)試的結(jié)果證明了本文對(duì)網(wǎng)絡(luò)安全管理系統(tǒng)改進(jìn)設(shè)計(jì)的合理性;而性能方面的測(cè)試數(shù)據(jù)表明改進(jìn)后的模塊性能得到了顯著的提升。(4)對(duì)內(nèi)存數(shù)據(jù)庫(kù)應(yīng)用于具體項(xiàng)目中的幾種方式進(jìn)行了歸納,總結(jié)出了三類對(duì)內(nèi)存數(shù)據(jù)庫(kù)的典型實(shí)踐,具有一定的借鑒意義。
[Abstract]:Because the memory database stores the data in the memory, compared with the traditional disk database, the performance has been greatly improved, and with the increasing of memory capacity and the decreasing of memory price, In recent years, the research on memory database has attracted more and more attention. Concurrency control and other research results have been emerging, memory database research has become a hot topic in database research. At the same time, in order to seize the market, the major database manufacturers have introduced related memory database products, which has accelerated the popularity of memory database applications. Nowadays, memory database has become one of the most important solutions to the problem of storage and retrieval performance. This paper takes the memory database as the research object, proposes a set of new algorithm of the index of the memory database, and takes the network security management system as the application scene, introduces the memory database and carries on the improvement design. The overall performance of the system is improved significantly. The main results are as follows: (1) A new index algorithm of DLB tree is proposed. By extending the leaf node of B tree to double leaf node, the related operation algorithm is improved. Finally, an index algorithm which is more suitable for range query and has higher space utilization is obtained. Experimental results show that the range query efficiency is twice that of B-tree query efficiency when the query data reaches a certain amount. (2) the memory database is applied to the network security management system. To improve the current system in the database storage and retrieval of the functional and performance defects, using the memory database in storage flexibility and retrieval efficiency, The single sign-on and privilege control function of the user management module and the event matching and real-time event monitoring of the security event management module are redesigned and improved. The corresponding persistence scheme is designed for the data inconsistency caused by the memory database. (3) the function and performance of the improved network security management system are tested. The result of function test proves the rationality of improving the design of network security management system in this paper. The performance test data show that the improved module performance has been significantly improved. (4) the application of memory database in specific projects is summarized, and three kinds of typical practice of memory database are summarized. Has certain reference significance.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP311.13;TP393.08

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 馬豫星;;Redis數(shù)據(jù)庫(kù)特性分析[J];物聯(lián)網(wǎng)技術(shù);2015年03期

2 董紹嬋;周敏奇;張蓉;周傲英;;內(nèi)存數(shù)據(jù)索引:以處理器為核心的性能優(yōu)化技術(shù)[J];華東師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2014年05期

3 張?jiān)?;內(nèi)存數(shù)據(jù)庫(kù)技術(shù)在金融信息服務(wù)平臺(tái)關(guān)鍵項(xiàng)目建設(shè)中的應(yīng)用及未來(lái)項(xiàng)目應(yīng)用研究[J];中國(guó)傳媒科技;2014年11期

4 楊運(yùn)平;吳智俊;;Apache Shiro安全框架在技術(shù)轉(zhuǎn)移服務(wù)系統(tǒng)中的應(yīng)用[J];計(jì)算機(jī)與現(xiàn)代化;2014年03期

5 徐寶磊;王安志;;基于Session的單點(diǎn)登錄系統(tǒng)研究與設(shè)計(jì)[J];重慶文理學(xué)院學(xué)報(bào)(自然科學(xué)版);2012年02期

6 張華強(qiáng);;關(guān)系型數(shù)據(jù)庫(kù)與NoSQL數(shù)據(jù)庫(kù)[J];電腦知識(shí)與技術(shù);2011年20期

7 許麗花;翁英萍;;Oracle Timesten內(nèi)存數(shù)據(jù)庫(kù)自動(dòng)數(shù)據(jù)清理機(jī)制探究[J];電腦知識(shí)與技術(shù);2010年35期

8 黃賢立;;NoSQL非關(guān)系型數(shù)據(jù)庫(kù)的發(fā)展及應(yīng)用初探[J];福建電腦;2010年07期

9 李蔚;張效尉;李剛;;內(nèi)存數(shù)據(jù)庫(kù)應(yīng)用于電信計(jì)費(fèi)系統(tǒng)的研究與實(shí)現(xiàn)[J];鄭州輕工業(yè)學(xué)院學(xué)報(bào)(自然科學(xué)版);2008年03期

10 王英強(qiáng);石永生;;B+樹在數(shù)據(jù)庫(kù)索引中的應(yīng)用[J];長(zhǎng)江大學(xué)學(xué)報(bào)(自然科學(xué)版)理工卷;2008年01期

相關(guān)碩士學(xué)位論文 前5條

1 崔解賓;分布式內(nèi)存緩存技術(shù)在數(shù)據(jù)處理平臺(tái)中的研究與應(yīng)用[D];北京郵電大學(xué);2015年

2 楊朝輝;主存數(shù)據(jù)庫(kù)索引機(jī)制的研究與改進(jìn)[D];南京航空航天大學(xué);2011年

3 葉飛;網(wǎng)絡(luò)安全管理系統(tǒng)的研究[D];西安電子科技大學(xué);2007年

4 何為超;網(wǎng)絡(luò)安全管理系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[D];西北工業(yè)大學(xué);2006年

5 王晨;內(nèi)存數(shù)據(jù)庫(kù)若干關(guān)鍵技術(shù)研究[D];浙江大學(xué);2006年

，

本文編號(hào)：2422105