【正文】 裝、維修、調(diào)式時,通過頭盔顯示器,可以將原來不能呈現(xiàn)的機(jī)器內(nèi)部結(jié)構(gòu),以及它的相關(guān)信息、數(shù)據(jù)完全呈現(xiàn)出來.現(xiàn)代的體感技術(shù),如微軟的Kinect以及Leap公司的Leap Motion體感控制器,能夠檢測和感知到人體的動作及手勢,進(jìn)而將動作轉(zhuǎn)化為對電腦及系統(tǒng)的控制,使人們擺脫了鍵盤、鼠標(biāo)、遙控器等傳統(tǒng)交互設(shè)備的束縛,如Google眼鏡,則有機(jī)地結(jié)合了大數(shù)據(jù)技術(shù)、增強(qiáng)現(xiàn)實、,我們可以實時地感知我們周圍的現(xiàn)實環(huán)境,并且通過大數(shù)據(jù)搜索、計算,實現(xiàn)對周遭的建筑、商家、人群、物體的實時識別和數(shù)據(jù)獲取,并疊加投射在我們的視網(wǎng)膜上,這樣可以實時地幫助我們工作、購物、休閑等,我們處在一個隨時被監(jiān)控、隱私被刺探、侵犯的狀態(tài),所以大數(shù)據(jù)技術(shù)所帶來的安全性問題也不容忽視.5 相關(guān)研究與我們的工作大數(shù)據(jù)的規(guī)模效應(yīng)給數(shù)據(jù)存儲管理以及數(shù)據(jù)分析帶來了極大的挑戰(zhàn),數(shù)據(jù)管理方式上的變革正在醞釀和發(fā)生,孟曉峰等學(xué)者對大數(shù)據(jù)的基本概念進(jìn)行剖析并對大數(shù)據(jù)的主要應(yīng)用作簡單對比,闡述并分析了大數(shù)據(jù)處理的基本框架和云計算技術(shù)對大數(shù)據(jù)時代數(shù)據(jù)管理所產(chǎn)生的作用,同時歸納總結(jié)大數(shù)據(jù)時代所面臨的新挑戰(zhàn)[49].陶雪嬌等人[51]介紹并分析了大數(shù)據(jù)的相關(guān)概念、特點(diǎn)、,有些學(xué)者指出面對數(shù)據(jù)處理的實時性有效性需求,我們需要根據(jù)大數(shù)據(jù)特點(diǎn)對傳統(tǒng)的常規(guī)數(shù)據(jù)處理技術(shù)進(jìn)行技術(shù)變革,形成適用于大數(shù)據(jù)收集、存儲、管理、處理、分析、共享和可視化的技術(shù)[52].上述綜述性論文更加注重于分析大數(shù)據(jù)技術(shù)的特點(diǎn)和發(fā)展趨勢,對大數(shù)據(jù)技術(shù)面臨的問題和分類介紹概括不夠完善.大數(shù)據(jù)分析相比于傳統(tǒng)的數(shù)據(jù)倉庫應(yīng)用,具有數(shù)據(jù)量大、查詢分析復(fù)雜等特點(diǎn),從大數(shù)據(jù)分析和數(shù)據(jù)倉庫架構(gòu)設(shè)計角度,文獻(xiàn)[33]首先列舉了大數(shù)據(jù)分析平臺需要具備的幾個重要特性,并對當(dāng)前的主流實現(xiàn)平臺并行數(shù)據(jù)庫、MapReduce及基于兩者的混合架構(gòu)行了分析歸納,指出了各自的優(yōu)勢及不足,HadoopDB[59][60],分析二者在發(fā)展過程中遇到的挑戰(zhàn)并指出關(guān)系數(shù)據(jù)管理技術(shù)和非關(guān)系數(shù)據(jù)管理技術(shù)在不斷的競爭中互相取長補(bǔ)短,在新的大數(shù)據(jù)分析生態(tài)系統(tǒng)內(nèi)找到自己的位置[55][58].在NoSQL系統(tǒng)的研究上,申德榮等[56]學(xué)者系統(tǒng)性總結(jié)了NoSQL系統(tǒng)的相關(guān)研究,包括體系結(jié)構(gòu)、數(shù)據(jù)模型、訪問方式、索引技術(shù)、事務(wù)特性、系統(tǒng)彈性、動態(tài)負(fù)載均衡、副本策略、數(shù)據(jù)一致性策略、基于flash的多級緩存機(jī)制、,分析不同的存儲策略和優(yōu)缺點(diǎn),缺少對大數(shù)據(jù)技術(shù)的全面性闡述,忽略了不同大數(shù)據(jù)技術(shù)之間以及大數(shù)據(jù)技術(shù)與云計算的協(xié)同作用.、模擬數(shù)據(jù)、傳感器數(shù)據(jù)和衛(wèi)星數(shù)據(jù)等“數(shù)據(jù)泛濫”問題[1],數(shù)據(jù)規(guī)模、(SWFMS)為科學(xué)計算提供了如數(shù)據(jù)管理、任務(wù)相關(guān)性、作業(yè)調(diào)度與執(zhí)行、[65],Kepler[63],Vistrails[64],Pegasus[62],Swift[39],VIEW[66]等工作流系統(tǒng)在許多領(lǐng)域都有廣泛的應(yīng)用,如物理學(xué)、天文學(xué)、生物信息學(xué)、神經(jīng)科學(xué)、,[67],來處理日益增長的數(shù)據(jù)量和分析復(fù)雜度,擁有大規(guī)模數(shù)據(jù)中心資源池和按需資源分配功能的云計算可以為科學(xué)工作流系統(tǒng)提供比上述環(huán)境更好的服務(wù),使工作流系統(tǒng)能夠處理PB級的科學(xué)問題.6 結(jié)束語、互聯(lián)網(wǎng)以及移動通信網(wǎng)絡(luò)的飛速發(fā)展催生了大數(shù)據(jù)問題,帶來了速度、結(jié)構(gòu)、體量、成本、價值、安全隱私、互聯(lián)互通等各方面的問題,傳統(tǒng)的信息技術(shù)處理手段在面對大數(shù)據(jù)問題時顯得力不從心,而同時大數(shù)據(jù)又能促進(jìn)云計算技術(shù)的真正落地和實施,、數(shù)據(jù)采集、數(shù)據(jù)存儲、數(shù)據(jù)計算以及數(shù)據(jù)展示與交互等方面描述大數(shù)據(jù)所涵蓋的幾大類技術(shù),從另一個角度為相關(guān)領(lǐng)域的學(xué)者描述大數(shù)據(jù)技術(shù)的挑戰(zhàn)與機(jī)遇,不斷的影響著我們的生活習(xí)慣和方式.致謝 在此,我們向?qū)Ρ疚牡墓ぷ鹘o予支持和建議的同行,尤其是電子科技大學(xué)計算機(jī)科學(xué)與工程學(xué)院極限網(wǎng)絡(luò)計算與服務(wù)實驗室中的同學(xué)和老師表示感謝.References:[1] Bell G, Hey T, Szalay A. Beyond the data deluge[J]. Science, 2009, 323(5919): 12971298.[2] Manyika J, Chui M, Brown B, Bughin J, Dobbs R, Roxburgh C, Byers AH. Big data: The next frontier for innovation, petition, and productivity. May 2011[J]. MacKinsey Global Institute, 2011.[3] Big Data Research and Development Initiative, [4] [5] Foster I, Zhao Y, Raicu I, Shiyong L. Cloud puting and grid puting 360degree pared[C]//Grid Computing Environments Workshop, 2008. GCE39。08. Ieee, 2008: 110.[6] OpenNebula, . [7] OpenNebula Architecture, : archives::architecture.[8] Openstack, .[9] Keahey K, Freeman T. Contextualization: Providing oneclick virtual clusters[C]//eScience, 2008. eScience39。08. IEEE Fourth International Conference on. IEEE, 2008: 301308.[10] Barham P, Dragovic B, Fraser K, Hand S, Harris T, Ho A, Neugebauer R, Pratt I, Warfield A. Xen and the art of virtualization[J]. ACM SIGOPS Operating Systems Review, 2003, 37(5): 164177.[11] KVM (Kernel Based Virtual Machine). [12] Nurmi D, Wolski R, Grzegorczyk C, Obertelli G, Soman S, Youseff L, Zagorodnov D. The eucalyptus opensource cloudputing system[C]//Cluster Computing and the Grid, 2009. CCGRID39。09. 9th IEEE/ACM International Symposium on. IEEE, 2009: 124131.[13] Ghemawat S, Gobioff H, Leung ST. The Google file system. In: Proc. of the 19th ACM Symp. on Operating Systems Principles. New York: ACM Press, 2003. 29?43.[14] Chang F, Dean J, Ghemawat S, Hsieh WC, Wallach DA, Burrows M, Chandra T, Fikes A, Gruber RE. Bigtable: A distributed storage system for structured data. In: Proc. of the 7th USENIX Symp. on Operating Systems Design and Implementation. Berkeley: USENIX Association, 2006. 205?218.[15] Zheng QL, Fang M, Wang S, Wang XQ, Wu XW, Wang H. Scientific Parallel Computing Based on MapReduce Model. Micro Electronics amp。 Computer, 2009, 26(8):1317 (in Chinese with English abstract).[16] Li GJ, Cheng XQ. Research Status and Scientific Thinking of Big Data[J]. Bulletin of Chinese Academy of Sciences, 2012, 27(6): 647657 (in Chinese with English abstract).[17] Thusoo A, Sarma J S, Jain N, Shao Z, Chakka P, Anthony S, Liu H, Wyckoff P, Murthy R. Hive: a warehousing solution over a mapreduce framework[J]. Proceedings of the VLDB Endowment, 2009, 2(2): 16261629.[18] Dean J, Ghemawat S. MapReduce: simplified data processing on large clusters[J]. Communications of the ACM, 2008, 51(1): 107113.[19] Gopalakrishna K, Hu G, Seth P. Communication layer using ZooKeeper. Yahoo! Inc., Tech. Rep., 2009.[20] Malewicz G, Austern MH, Bik AJC, Dehnert JC, Horn I, Leiser N, Czajkowski G. Pregel: a system for largescale graph processing[C]//Proceedings of the 2010 ACM SIGMOD International Conference on Management of data. ACM, 2010: 135146.[21] Olston C, Reed B, Srivastava U, Kumar R, Tomkins A. Pig latin: a notsoforeign language for data processing[C]//Proceedings of the 2008 ACM SIGMOD international conference on Management of data. ACM, 2008: 10991110.[22] Malkin J, Schroedl S, Nair A, Neumeyer L. Tuning Hyperparameters on Live Traffic with S4. In TechPulse 2010: Internal Yahoo! Conference, 2010.[23] Schroedl S, Kesari A, Neumeyer L. Personalized ad placement in web search[C]//Proceedings of the 4th Annual International Workshop on Data Mining and Audience Intelligence for Online Advertising (AdKDD), Washington USA. 2010.[24] Stonebraker M, 199。etintemel U, Zdonik S. The 8 requirements of realtime stream processing[J]. ACM SIGMOD Record, 2005, 34(4): 4247.[25] Apache Hadoop. [26] Neumeyer L, Robbins B, Nair A, Kesari A. S4: Distributed stream puting platform[C]//Data Mining Workshops (ICDMW), 2010 IEEE International Conference on. IEEE, 2010: 170177.[27] Khetrapal A, Ganesh V. HBase and Hypertable for large scale distributed storage systems[J]. Dept. of Computer Science, Purdue University, 2006.[28] [29] [30] [31] Li YL, Dong J. Study and Improvement of MapReduce based on Hadoop. Computer Engineering and Design. 2012, 33(8):31103116 (in Chinese with English abstract). [32] Baker J, Bond C, Corbett JC, Furman JJ, Khorlin A, Larson J, Leon JM, Li YW, Lloyd A, Yushprakh V. Megastore: Providing Scalable, Highly Available Storage for Interactive S