【正文】 to verify and validate substitutions, etc), extensibility and scalability(ability to readily extend system ponent pool and to scale it, increase capabilities of individual ponents, have an extensible and scalable architecture that can automatically discover new functionalities and resources, etc), customizability (ability to customize generic features to the needs of a particular scientific domain and problem), and posability (easy construction of more plex functional solutions using basic ponents, reasoning about such positions, etc.). There are other characteristics that also are very important. Those include reliability and availability of the ponents and services, the cost of the services, security, total cost of ownership, economy of scale, and so on. In the context of cloud puting we distinguish many categories of ponents. From differentiated and undifferentiated hardware, to generalpurpose and specialized software and applications, to real and virtual ―images‖, to environments, to noroot differentiated resources, to workflowbased environments and collections of services, and so on. They are discussed later in the paper. 3 . Workflows An integrated view of servicebased activities is provided by the concept of a workflow. An ITassisted workflow represents a series of structured activities and putations that arise in information assisted problemsolving. Workflows have been drawing enormous attention in the database and information systems research and development munities [., Geo95, Hsu93]. Similarly, the scientific munity has developed a number of problemsolving environments, most of them as integrated solutions [Hou00]. Scientific workflows merge advances in these two areas to automate support for sophisticated scientific problemsolving [., Lud06, Vou97]. A workflow can be represented by a directed graph that represents dataflows that connect loosely and tightly coupled (and often anonymous) processing ponents. One such graph is shown in Figure 1. It illustrates a Keplerbased implementation of a part of a fusion simulation workflow [Alt07a, Bat07]. Figure 1. A Keplerbased workflow. In the context of ―cloud puting‖, the key questions should be whether the underlying infrastructure is supportive of the workfloworiented view of the world. This includes ondemand and advancereservation based access to individual and aggregated putational and other resources, autonomics, ability to group resources from potentially different ―clouds‖ to deliver workflow results, appropriate level of security and privacy, etc. . Virtualization Virtualization is another very useful concept. It allows abstraction and isolation of lowerlevel functionalities and underlying hardware. This enables portability of higherlevel functions and sharing and/or aggregation of the physical resources. The virtualization concept has been around in some form since 1960s (., in IBM mainframe systems). Since then, the concept has matured considerably and it has been applied to all aspects of puting – memory, storage, processors, software, works, as well as services that IT offers. It is the bination of the growing needs and the recent advances in the IT architectures and solutions that is now bringing the virtualization to the true modity level. Virtualization, through its economy of scale, and its ability to offer very advanced and plex IT services at a reasonable cost, is poised to bee, along with wireless and highly distributed and pervasive puting devices, such as sensors and personal cellbased access devices, the driving technology behind the next waive in IT growth [Vou08b]. Not surprisingly there are dozens of virtualization products, and a number of small and large panies that make them. Some examples in the operating systems and 4 software applications space are VMware1, Xen an open source Linuxbased product developed by XenSource2, and Microsoft virtualization products3, to mention a few. Major IT players have also shown a renewed interest in the technology4,5,6,7,8,9 [IBM06, Sun06]. Classical storage players such as EMC10, NetApp11, IBM12 and Hitachi13 have not been standing still either. In addition, the work virtualization market is teeming with activity. . Users The most important Cloud entity, and the principal quality driver and constraining influence is, of course, the user. The value of solutions depends very much on the view it has of its enduser requirements and user categories. Figure 2. Cloud user hierarchy 1 2 3 4 webamp。 [Wik08]百科， “ 云計(jì)算 ” 2020年5月。知識工程，與 vol 11（ 4）， 7月 / 1999年 8月，第673687 。 [Sun06]保羅磷蝦， Sun公司的 Solaris 獲得安全，虛擬化提高了， “ 信息世界 ”，12/12/2020 。scher，一 Altintas，角伯克利，希金斯四，五積弗蘭克，米瓊斯，E 16（ 2）， 1993年 6月。 [Den96]丹尼斯 .， ， ，光 j 的 Galluppi， ，馬 Vouk，|“ 空氣質(zhì)量的綜合建模下一代：環(huán)保局的模型 3” 大氣環(huán)境， 30卷（ 12），頁 1925至 1938年， 1996年。 116 [Bar07]羅斯里納巴雷托，特倫斯克里奇洛，艾拉汗，斯科特 Klasky， Leena科拉，杰弗里利根，皮埃爾 Mouallem， Meiyappan Nagappan，諾伯特 Podhorszki，姆拉登納萊 Vouk， ―科學(xué)的工作流程管理和監(jiān)控，通過儀表板， ‖海報(bào)＃ 93，微軟 eScience 研討會星期五中心，北卡羅萊納大學(xué)查珀?duì)柹奖笨_來納州， 10月 13 15,2020，第 108。 6 致謝 我要感謝在北卡羅來納州的 VCL 球隊(duì)做出的貢獻(xiàn)我的同事們的見解和支持。有趣的是，看到有可能是一個 機(jī)會，也許是在上午 12點(diǎn)轉(zhuǎn)移到上午 07點(diǎn)時間段的一些高性能計(jì)算資源，這是 9 永遠(yuǎn)的 CPU 時間短，目前尚不清楚，這可能是一種節(jié)約成本的措施。 圖 7 VCL的每日變化在單席的利用率（超過四年的平均值）。雖然北卡州 立經(jīng)驗(yàn)與優(yōu)秀的 VCL和我們的安全解決方案已被高舉在過去四年精美，安全往往是一個移動的目標(biāo)和挑戰(zhàn)。 還有一個圖像的便攜性和暗示圖像格式的問題。這是有相當(dāng)數(shù)量的基本 的元數(shù)據(jù)，一些永久連接到一個圖像，動態(tài)地連接到一個圖像，一些在云數(shù)據(jù)庫管理中存儲。從功能上看，它已載入了以分布式計(jì)算為基礎(chǔ)的圖像葉片數(shù) [AEC08]，大路口 [Had08]一個可以實(shí)現(xiàn)谷歌般的地圖 /減少環(huán)境中，通過加載和環(huán)境組成的集團(tuán)或球，可以構(gòu)造一個子云電網(wǎng)為基礎(chǔ)計(jì)算等的圖像 。更為詳細(xì)的 VCL用戶服務(wù)，功能，安全性和概念在信息中 可以找到 [Ave07， Vou08a]。 7 圖 5 說明了北卡羅萊納州立云上的 VCL技術(shù)。但是，已成立一個較低的 VCL的終端到終端連接的吞吐量， DSL 和電纜調(diào)制解調(diào)器速度的約束。它可以根據(jù)需要啟動任何以前的資源數(shù)量，并管理其使用和釋放一個自動化的工作流程為基礎(chǔ)。需要可能從 “ 裸機(jī) ” 裝的影像，（在管理程序）虛擬平臺圖像，圖像總量集合（環(huán)境），以圖像與一些限制，以工作流為基礎(chǔ)的服務(wù)。 程序設(shè)計(jì)者必須是組件（基準(zhǔn)線圖像和應(yīng)用程序）的專家，而且必須具備上述這些圖中的兩個三角形的用戶類別的需要很好地理解。它集成以下內(nèi)容： a）任何基線的操作系統(tǒng)，如