云計(jì)算（cloud computing，臺(tái)灣譯作云端運(yùn)算），是分布式計(jì)算技術(shù)的一種，其最基本的概念，是透過(guò)網(wǎng)絡(luò)將龐大的計(jì)算處理程序自動(dòng)分拆成無(wú)數(shù)個(gè)較小的子程序，再交由多部服務(wù)器所組成的龐大系統(tǒng)經(jīng)搜尋、計(jì)算分析之后將處理結(jié)果回傳給用戶。透過(guò)這項(xiàng)技術(shù)，網(wǎng)絡(luò)服務(wù)提供者可以在數(shù)秒之內(nèi)，達(dá)成處理數(shù)以千萬(wàn)計(jì)甚至億計(jì)的信息，達(dá)到和“超級(jí)計(jì)算機(jī)”同樣強(qiáng)大效能的網(wǎng)絡(luò)服務(wù)。

云計(jì)算是一種資源交付和使用模式，指通過(guò)網(wǎng)絡(luò)獲得應(yīng)用所需的資源（硬件、平臺(tái)、軟件）。提供資源的網(wǎng)絡(luò)被稱為“云”?！霸啤敝械馁Y源在使用者看來(lái)是可以無(wú)限擴(kuò)展的，并且可以隨時(shí)獲取。這種特性經(jīng)常被比喻為像水電一樣使用硬件資源，按需購(gòu)買和使用。^[1]（Cloud computing is a resource delivery and usage model, it means get resource (Hardware, software)via network. The network of providing resource is called ‘Cloud’. The hardware resource in the ‘Cloud’ seems scalable infinitely and can be used whenever.）^[2]

最簡(jiǎn)單的云計(jì)算技術(shù)在網(wǎng)絡(luò)服務(wù)中已經(jīng)隨處可見，例如搜尋引擎、網(wǎng)絡(luò)信箱等，使用者只要輸入簡(jiǎn)單指令即能得到大量信息。

未來(lái)如手機(jī)、GPS等行動(dòng)裝置都可以透過(guò)云計(jì)算技術(shù)，發(fā)展出更多的應(yīng)用服務(wù)。

進(jìn)一步的云計(jì)算不僅只做資料搜尋、分析的功能，更可計(jì)算一些像是分析DNA結(jié)構(gòu)、基因圖譜定序、解析癌癥細(xì)胞等。

稍早之前的大規(guī)模分布式計(jì)算技術(shù)即為“云計(jì)算”的概念起源^[1]。

Google目前的云技術(shù)，主要由MapReduce、GFS及BigTable三項(xiàng)所組成

云是指因特網(wǎng)，因過(guò)去一直將因特網(wǎng)畫成一朵云。

云端儲(chǔ)存（cloud storage）是一種將數(shù)據(jù)保存在虛擬服務(wù)器上的數(shù)據(jù)類型，通常意義上，數(shù)據(jù)存儲(chǔ)在第三方媒介，而非特定單一服務(wù)器上。

2007年10月，Google與IBM開始在美國(guó)大學(xué)校園，包括卡內(nèi)基美隆大學(xué)、麻省理工學(xué)院、史丹佛大學(xué)、加州大學(xué)柏克萊分校及馬里蘭大學(xué)等，推廣云計(jì)算的計(jì)劃，這項(xiàng)計(jì)劃希望能降低分布式計(jì)算技術(shù)在學(xué)術(shù)研究方面的成本，并為這些大學(xué)提供相關(guān)的軟硬件設(shè)備及技術(shù)支援（包括數(shù)百臺(tái)個(gè)人計(jì)算機(jī)及BladeCenter與System x服務(wù)器，這些計(jì)算平臺(tái)將提供1600個(gè)處理器，支援包括Linux、Xen、Hadoop等開放源代碼平臺(tái)）。而學(xué)生則可以透過(guò)網(wǎng)絡(luò)開發(fā)各項(xiàng)以大規(guī)模計(jì)算為基礎(chǔ)的研究計(jì)劃^[1]。

2008年1月30日，Google宣布在臺(tái)灣啟動(dòng)“云計(jì)算學(xué)術(shù)計(jì)劃”，將與臺(tái)灣大學(xué)、臺(tái)灣交通大學(xué)等學(xué)校合作，將這種先進(jìn)的大規(guī)模、快速計(jì)算技術(shù)推廣到校園^[3]。

2008年8月3日，美國(guó)專利商標(biāo)局(以下簡(jiǎn)稱“SPTO”)網(wǎng)站信息顯示，戴爾正在申請(qǐng)“云計(jì)算”(Cloud Computing)商標(biāo)，此舉旨在加強(qiáng)對(duì)這一未來(lái)可能重塑技術(shù)架構(gòu)的術(shù)語(yǔ)的控制權(quán)。戴爾在申請(qǐng)文件中稱，云計(jì)算是“在數(shù)據(jù)中心和巨型規(guī)模的計(jì)算環(huán)境中，為他人提供計(jì)算機(jī)硬件定制制造”。^[4]。

云計(jì)算是隨著處理器技術(shù)、虛擬化技術(shù)、分布式存儲(chǔ)技術(shù)、寬帶互聯(lián)網(wǎng)技術(shù)和自動(dòng)化管理技術(shù)的發(fā)展而產(chǎn)生的. 這種大規(guī)模的計(jì)算能力通常是由分布式的大規(guī)模集群和服務(wù)器虛擬化軟件搭建。(Cloud computing and technology:New advances in processors, virtualization technology, distributed storage, broadband Internet access , automated management and fast, inexpensive servers have all combined to make cloud computing a compelling paradigm.This vast process power is usually got with a distributed, large-scale server cluster and server virtualization software.)

傳統(tǒng)的計(jì)算模式下，單臺(tái)臺(tái)式機(jī)的資源用來(lái)完成任務(wù)。在客戶服務(wù)器模式下，服務(wù)器用來(lái)執(zhí)行任務(wù)。在云計(jì)算模式下，網(wǎng)絡(luò)超級(jí)計(jì)算機(jī)—“云”用來(lái)執(zhí)行任務(wù)。用戶能在任何時(shí)間任何地點(diǎn)通過(guò)互聯(lián)網(wǎng)獲取計(jì)算、存儲(chǔ)、網(wǎng)絡(luò)資源，并且能夠按照處理器利用率、存儲(chǔ)使用量、帶寬消耗付費(fèi)。(Cloud computing usage model:In traditional computing model, tasks are completed using a single desktop computer. In the client/serer model, tasks are completed using a remote server. In cloud computing model, tasks are completed using ‘Cloud’-a network super computer.Cloud computing allows users and companies to pay for and use the services and storage that they need, when they need them and, as wireless broadband connection options grow, where they need them. Customers can be billed based upon server utlilization, processing power used or bandwidth consumed.)

云計(jì)算有可能顛覆軟件產(chǎn)業(yè)，應(yīng)用和許可被隨時(shí)購(gòu)買和生效，應(yīng)用在網(wǎng)絡(luò)上而不是本機(jī)上運(yùn)行。這種轉(zhuǎn)變將數(shù)據(jù)中心放在網(wǎng)絡(luò)的核心位置，而所有的應(yīng)用所需要的計(jì)算能力、存儲(chǔ)、帶寬、電力都由數(shù)據(jù)中心提供。云計(jì)算不僅影響商業(yè)模式，還影響開發(fā)、部署、運(yùn)行、交付應(yīng)用的方式。(Cloud computing impact:As a result, cloud computing has the potential to upend the software industry entirely, as applications are purchased, licensed and run over the network instead of a user's desktop. This shift will put data centers and their administrators at the center of the distributed network, as processing power, electricity, bandwidth and storage are all managed remotely. It affects not only business models, but the underlying architecture of how we develop, deploy, run and deliver applications.)^[5]

• 快速部署(Fast Provision)

• 縮小主機(jī)規(guī)模(Reduce servers scale)

• 提高資源利用率(Increase resource utilization rate)

• 提高管理效率(Improve management efficiency)

• 降低運(yùn)維成本(Lower maintenance cost)

• 基礎(chǔ)設(shè)施可以放置在低土地和能源成本的地區(qū)(Location of infrastructure in areas with lower costs of real estate and electricity)

• 提供商業(yè)連續(xù)性服務(wù)(Provide business continuity service)

• 提高管理效率(Improve management efficiency)

• 提高服務(wù)水平(Improve service levels)

• 復(fù)雜的體系結(jié)構(gòu)(Complex architecture)

• 商業(yè)模式和理念的轉(zhuǎn)變(Change of business model and faith)

• 用戶端負(fù)載降低(Lower client workload)

• 降低總體擁有成本(Lower Total Cost Ownership)

• 可能將應(yīng)用的開發(fā)與基礎(chǔ)設(shè)施維護(hù)相對(duì)分離(Separation of infrastructure maintenance duties from domain-specific application development)

• 可能將程序代碼與物理資源分離(Separation of application code from physical resources)

• 不需要為一次性任務(wù)或罕見的負(fù)載狀況準(zhǔn)備大量設(shè)備(Not have to purchase assets for one-time or infrequent intensive computing tasks)

• 按需擴(kuò)展資源(Expand resource on-demand)

• 使應(yīng)用具有高可用性(Make the application have high availability)

• 快速部署應(yīng)用(Quickly deploy application)

• 按使用付費(fèi)(Pay per use)

• 自愈合(Self-healing)

• 多用戶使用(Multi-tenancy)

• 虛擬化(Virtualized)

• 線形擴(kuò)展(Linearly ScalableLinearly Scalable)

• 資源監(jiān)控和測(cè)量(Resource Monitor and measure)

• 資源注冊(cè)和發(fā)現(xiàn)(Resource registration and discovery)

云計(jì)算正成為行業(yè)中的熱點(diǎn)概念。它和下面的詞匯產(chǎn)生了或多或少的關(guān)連：網(wǎng)格計(jì)算，效用計(jì)算，虛擬化，服務(wù)器集群，主機(jī)租用，主機(jī)托管。云計(jì)算平臺(tái)普遍用到了虛擬化技術(shù)，并且基于一個(gè)集群來(lái)構(gòu)建，和網(wǎng)格計(jì)算和效用計(jì)算以及SaaS有著天然的聯(lián)系，而且被用在IDC行業(yè)與主機(jī)租用和主機(jī)托管業(yè)務(wù)競(jìng)爭(zhēng)。(Cloud computing’s brother buzzwords:Cloud computing is becoming one of the next industry buzz words. And it has more or less relation with these words: grid computing, utility computing, virtualization, server cluster, Dedicated server, Colocation. Cloud computing infrastructure usually use virtualization technology, and is built based on a server cluster, have nature relation to grid computing and utility computing, and is use to compete with Dedicated server and Colocation.)

云計(jì)算異軍突起

相對(duì)與云計(jì)算的兄弟概念而言，云計(jì)算只是剛剛興起。從Google Trend 獲取的信息我們可以看到，云計(jì)算在2007年末才被該系統(tǒng)統(tǒng)計(jì)，并且一直處于上升趨勢(shì)。而同時(shí)，網(wǎng)格計(jì)算、效用計(jì)算、分布式計(jì)算的概念卻呈現(xiàn)下降趨勢(shì)。云計(jì)算和虛擬化應(yīng)該會(huì)在托管平臺(tái)上有所作為，而主機(jī)租用已經(jīng)呈現(xiàn)下降趨勢(shì)。(Cloud computing break out:Compared to its brother buzzwords, cloud computing is just beginning. Trends in usage of the terms from Google searches shows Cloud Computing is a relatively new term introduced in the past year. There has also been a decline in general interest of Grid, Utility and Distributed computing. Cloud Computing and Virtualization are the next hot hosting platforms; the Dedicated server term is slowly starting to lose ground vs. Virtualization and Cloud Computing.)

云計(jì)算和效用計(jì)算的比較

效用計(jì)算是一種提供計(jì)算資源的商業(yè)模式，用戶從計(jì)算資源供應(yīng)商獲取和使用計(jì)算資源并基于實(shí)際使用的資源付費(fèi)。簡(jiǎn)單說(shuō)，是一種基于資源使用量的付費(fèi)模式。效用計(jì)算主要給用戶帶來(lái)經(jīng)濟(jì)效益。企業(yè)數(shù)據(jù)中心的資源利用率普遍在20%左右，這主要是因?yàn)槌~部署—購(gòu)買比平均所需資源更多的硬件一邊處理峰值負(fù)載，可預(yù)計(jì)到的或不可預(yù)計(jì)的。效用計(jì)算則允許用戶只為他們所需要用到并且已經(jīng)用到的那部分資源付費(fèi)。（Utility computing is a business model of providing computing resource， user get and use the computing resource from service provider and pay for practically used resource. To say it simply, it is a price model based on resource usage quantity. The main benefit of utility computing is better economics. Corporate data centers are notoriously underutilized, with resources such as servers often idle 85 percent of the time. This is due to overprovisioning — buying more hardware than is needed on average in order to handle peaks (such as the opening of the Wall Street trading day or the holiday shopping season), to handle expected future loads and to prepare for unanticipated surges in demand. Utility computing allows companies to only pay for the computing resources they need, when they need them.）

效用計(jì)算是是一種分發(fā)應(yīng)用所需資源的計(jì)費(fèi)模式。 云計(jì)算是一種計(jì)算模式， 代表了在某種程度上共享資源進(jìn)行設(shè)計(jì)、開發(fā)、部署、運(yùn)行應(yīng)用，以及資源的可擴(kuò)展收縮和對(duì)應(yīng)用連續(xù)性的支持。效用計(jì)算通常需要云計(jì)算基礎(chǔ)設(shè)施支持，但并不是一定需要。同樣，在云計(jì)算之上可以提供效用計(jì)算，也可以不采用效用計(jì)算。（Comparison of Utility Computing and Cloud Computing：Utility computing is a business model, it is a type of price model to deliver application infrastructure resource. Cloud computing is a computing model, relates to the way we design, build, deploy and run applications that operate in a sharing resources and boasting the ability to dynamically grow, shrink and self-heal. Utility computing is often need a cloud computing infrastructure, but not must need. Sameness, above the cloud computing, we can adopt utility computing, and, we can adopt other price model.）^[6][7][8]

分布式計(jì)算

分布式計(jì)算是指在一個(gè)松散或嚴(yán)格約束條件下使用一個(gè)硬件和軟件系統(tǒng)處理任務(wù)，這個(gè)系統(tǒng)包含多個(gè)處理器單元或存儲(chǔ)單元，多個(gè)并發(fā)的過(guò)程，多個(gè)程序。一個(gè)程序被分成多個(gè)部分，同時(shí)在通過(guò)網(wǎng)絡(luò)連接起來(lái)的計(jì)算機(jī)上運(yùn)行。分布式計(jì)算類似于并行計(jì)算，但并行計(jì)算通常用于指一個(gè)程序的多個(gè)部分同時(shí)運(yùn)行于某臺(tái)計(jì)算機(jī)上的多個(gè)處理器上。所以，分布式計(jì)算通常必須處理異構(gòu)環(huán)境、多樣化的網(wǎng)絡(luò)連接、不可預(yù)知的網(wǎng)絡(luò)或計(jì)算機(jī)錯(cuò)誤。(Distributed Computing:Distributed computing deals with hardware and software systems containing more than one processing element or storage element, concurrent processes, or multiple programs, running under a loosely or tightly controlled regime.In distributed computing， a program is split up into parts that run simultaneously on multiple computers communicating over a network. Distributed computing is a form of parallel computing, but parallel computing is most commonly used to describe program parts running simultaneously on multiple processors in the same computer. Both types of processing require dividing a program into parts that can run simultaneously, but distributed programs often must deal with heterogeneous environments, network links of varying latencies, and unpredictable failures in the network or the computers.)^[9][10]

云計(jì)算與網(wǎng)格計(jì)算的不同點(diǎn)

網(wǎng)格計(jì)算是指分布式計(jì)算中兩類比較廣泛使用的子類型。一類是，在分布式的計(jì)算資源支持下作為服務(wù)被提供的在線計(jì)算或存儲(chǔ)。另一類是，一個(gè)松散連接的計(jì)算機(jī)網(wǎng)絡(luò)構(gòu)成的一個(gè)虛擬超級(jí)計(jì)算機(jī)，可以用來(lái)執(zhí)行大規(guī)模任務(wù)。該技術(shù)通常被用來(lái)通過(guò)志愿者計(jì)算解決計(jì)算敏感型的科研、數(shù)學(xué)、學(xué)術(shù)問(wèn)題，也被商業(yè)公司用來(lái)進(jìn)行電子商務(wù)和網(wǎng)絡(luò)服務(wù)所需的后臺(tái)數(shù)據(jù)處理、經(jīng)濟(jì)預(yù)測(cè)、地震分析等。(Grid computing:Grid computing is a term for either of two broad subcategories of distributed computing: 1 Online computation or storage offered as a service supported by a pool of distributed computing resources, also known as utility computing, on-demand computing, or cloud computing. Data grids provide controlled sharing and management of large amounts of distributed data, often used in combination with computational grids. 2 The creation of a "virtual supercomputer" composed of a network of loosely-coupled computers, acting in concert to perform very large tasks. This technology has been applied to computationally-intensive scientific, mathematical, and academic problems through volunteer computing, and it is used in commercial enterprises for such diverse applications as drug discovery, economic forecasting, seismic analysis, and back-office data processing in support of e-commerce and web services.^[11]

網(wǎng)格計(jì)算強(qiáng)調(diào)資源共享，任何人都可以做為請(qǐng)求者使用其它節(jié)點(diǎn)的資源，任何人都需要貢獻(xiàn)一定資源給其他節(jié)點(diǎn)。網(wǎng)格計(jì)算強(qiáng)調(diào)將工作量轉(zhuǎn)移到遠(yuǎn)程的可用計(jì)算資源上。云計(jì)算強(qiáng)調(diào)專有，任何人都可以獲取自己的專有資源，并且這些資源是由少數(shù)團(tuán)體提供的，使用者不需要貢獻(xiàn)自己的資源。在云計(jì)算中，計(jì)算資源被轉(zhuǎn)換形式去適應(yīng)工作負(fù)載，它支持網(wǎng)格類型應(yīng)用，也支持非網(wǎng)格環(huán)境，比如運(yùn)行傳統(tǒng)或Web2.0應(yīng)用的三層網(wǎng)絡(luò)架構(gòu)。

網(wǎng)格計(jì)算側(cè)重并行的計(jì)算集中性需求，并且難以自動(dòng)擴(kuò)展。云計(jì)算側(cè)重事務(wù)性應(yīng)用，大量的單獨(dú)的請(qǐng)求，可以實(shí)現(xiàn)自動(dòng)或半自動(dòng)的擴(kuò)展。(Grid computing emphasizes on resource sharing, every grid node can apply for resource from other nodes, and every node should contribute resource to the grid. The focus of grid computing is on the ability of moving a workload to the location of theneeded computing resources, which are mostly remote and are readily available for use.Grids also require applications to conform to the grid software interfaces.Cloud computing emphasize on proprietary, every user out of the cloud can get it’s own private resource from the cloud, and the cloud resource are provided by the specific service provider, the user need not contribute its resource. In a cloud environment, computing resouces, such as servers, can be dynamically shaped or carved out from its underlying hardware infrastructure and made available to a workload. In addition, while a cloud does support grid, a cloud can also support nongrid environments,such as a three-tier Web architecture running traditional or Web 2.0 applications.Grid computing emphasizes on computing sensitive task, and is difficult to automated scale. Cloud computing emphasizes on transactional application, a great amount of separate request, and can scale automatically or semiautomatically.)^[12][13][14]

服務(wù)器集群

服務(wù)器集群是指將一組服務(wù)器關(guān)聯(lián)起來(lái)，使它們?cè)谕饨鐝暮芏喾矫婵雌饋?lái)如同一臺(tái)服務(wù)器。集群內(nèi)的服務(wù)器之間通常通過(guò)局域網(wǎng)連接，通常用來(lái)改善性能和可用性，但一般而言比具有同等性能功能和可用性的單臺(tái)主機(jī)具有更低的成本。(Computer cluster:A computer cluster is a group of coupled computers that work together closely so that in many respects they can be viewed as though they are a single computer. The components of a cluster are commonly, but not always, connected to each other through fast local area networks. Clusters are usually deployed to improve performance and/or availability over that provided by a single computer, while typically being much more cost-effective than single computers of comparable speed or availability.^[15])

網(wǎng)格通常更加松散連接、異構(gòu)、地理位置分散，主機(jī)之間信任度更低。(Grids tend to be more loosely coupled, heterogeneous, and geographically dispersed, grid computers do not fully trust each other.)

虛擬化

虛擬化指對(duì)計(jì)算資源進(jìn)行抽象的一個(gè)廣義概念。虛擬化對(duì)上層應(yīng)用或用戶隱藏了計(jì)算資源的底層屬性。它既包括使單個(gè)的資源（比如一個(gè)服務(wù)器，一個(gè)操作系統(tǒng)，一個(gè)應(yīng)用程序，一個(gè)存儲(chǔ)設(shè)備）劃分成多個(gè)虛擬資源，也包括將多個(gè)資源（比如存儲(chǔ)設(shè)備或服務(wù)器）整合成一個(gè)虛擬資源。虛擬化技術(shù)是指實(shí)現(xiàn)虛擬化的具體的技術(shù)性手段和方法的集合性概念。虛擬化技術(shù)根據(jù)對(duì)象可以分成存儲(chǔ)虛擬化、計(jì)算虛擬化、網(wǎng)絡(luò)虛擬化等。計(jì)算虛擬化可以分為操作系統(tǒng)級(jí)虛擬化，應(yīng)用程序級(jí)，和虛擬機(jī)管理器。虛擬機(jī)管理器分為宿主虛擬機(jī)和客戶虛擬機(jī)。(Virtualization:Virtualization is a broad term that refers to the abstraction of computer resources. Virtualization hides the physical characteristics of computing resources from their users, be they applications, or end users.^[15] This includes making a single physical resource (such as a server, an operating system, an application, or storage device) appear to function as multiple virtual resources; it can also include making multiple physical resources (such as storage devices or servers) appear as a single virtual resource.^[15] Virtualization technology is a aggregative term of technical means and methods to implement virtualization. It can be divided to many types based on objects: storage virtualization, computing virtualization, network virtualization. Computing virtualization include：OS level virtualization, application level virtualization, hyper visor. Hypervisor include: host vm and guest vm.)

• 連續(xù)高可用性(Continuous high availability)
  • 某個(gè)集群的失效處理

• 一致性(Consistency)
  • 不同集群的同步

• 互操作性和標(biāo)準(zhǔn)化(Interoperability and standarlization)
  • 在萌芽和成長(zhǎng)期，各廠商都試圖建立自己的接口API

• 所有構(gòu)件的擴(kuò)展(Scalability of all components)

• 信息保密(Data secrecy)

• 跨地區(qū)存儲(chǔ)和數(shù)據(jù)傳輸可能會(huì)引發(fā)法律和政治問(wèn)題(Legal and political problem of data store and translation across regions)

• 性能問(wèn)題(Performance issue)

• 差異化定制問(wèn)題(Difficulty customizing)

• 組織障礙(Organizational obstacle)

云計(jì)算平臺(tái)一般分為以下幾層：物理設(shè)施，虛擬化，管理，服務(wù)提供。物理設(shè)施被虛擬化，提供一個(gè)靈活的資源池體提高資源利用率。管理層負(fù)責(zé)物理資源和虛擬資源池的管理、部署、監(jiān)控、報(bào)警等。服務(wù)提供層組合管理層的功能提供某種形式的服務(wù)。(he physical hardware layer is virtualized to provide a flexible adaptive platform to improve resource utilization. The keys to new enterprise data center infrastructure services are the next two layers, the virtualization environment and management layer. The combination of these two layers ensure that resources in a data center are efficiently managed and can be provisioned, deployed, and configured rapidly.)

1. The NY Times(Amazon EC2)

2. Nasdaq(Amazon S3)

3. Major League Baseball(Joyent)

4. ESPN(Rightscale using Amazon EC2)

5. Hasbro(Amazon EC2)

6. British Telecom(3Tera)

7. Taylor Woodrow(Google Apps)

8. CSS(Amazon EC2)

9. Activision(Amazon EC2)

10. Business Objects (A SAP Company)(Rightscale using Amazon EC2)

云計(jì)算技術(shù)和方案提供者

• 3Tera - AppLogic grid OS used as cloud computing platform by service providers and enterprises

• Appistry - Cloud computing middleware - Enables easily scalable cloud computing in the enterprise.

• Cassatt - Cassatt Active Response platform enables administrators to set policies to power physical and virtual servers safely on and off and pool their computing resources.

• CloudScale Networks - Cloud enabler. Currently in private ALPHA only

• CloudHan - Cloud tech and infrastructure consultant, in China.

• Enomaly Inc - Service Provider & Cloud Enabler - Developer of the Enomalism Elastic Computing Platform & Elastic Drive

• Q-layer - provides software for data centers that enables cloud computing, support VSAN, VLAN, VPDC, currently support VMware ESX.

• Skytap - IaaS service optimized for QA, Training, Demo, and Ops Testing. Supports VMware, Xen hypervisors & Windows, Linux & Solaris OS guests.

云計(jì)算基礎(chǔ)設(shè)施層服務(wù)

• Agathon Group - Cloud provider. Services include highly available VPS, virtual private datacenters and ready-to-use LAMP stacks. Self-service ordering. Custom development and managed services available.

• Amazon Web Services - Amazon EC2/S3 (Hardware-a-a-S & Cloud Storage)

• CohesiveFT - CohesiveFT Elastic Server On-Demand

• ElasticHosts - UK-based instant, on-demand servers in the cloud

• Flexiscale - Another instant provisioner of web servers with some advanced features like auto-scaling coming soon.

• GoGrid - instant, on-demand servers offering "control in the cloud". Deploy Windows/Linux servers via web-interface in minutes

• GridLayer - Cloud Provider. A service by Layered Technologies that delivers Virtual Private Datacenters and virtual private servers from grids of commodity servers

• LayeredTechnologies - Cloud Provider. provider of on-demand hosting and cloud and utility computing solutions through its brand GridLayer

• Mosso - Rackspace's cloud hosting service

• Newservers - Instant provisioning of web servers either Windows or Linux

云計(jì)算平臺(tái)層服務(wù)

• Bungee Connect - Provides end to end tools and systems required to develop, deploy and host web applications (Platform as a Service)

• Coherence - Oracle Coherence Data Grid for EC2 and other cloud platforms

• Force.com - Salesforce.com's application development platform (PaaS)

• GigaSpaces - middleware for the cloud, "cloudware"

• Google AppEngine - (PaaS)Now support python

• Heroku - Ruby on Rails in their Cloud

• Qrimp - An AJAX based PaaS

• RightScale - RightScale provides a platform and expertise that enable companies to create scalable web applications running on Amazon’s Web Services that are reliable, easy to manage, and cost less

基于云計(jì)算的服務(wù)（Saas，云存儲(chǔ)）

• CAM Solutions - SaaS Provider. Cloud Event Management, Autonomics and Monitoring-as-a-Service(TM)

• CloudStatus- CloudEnabler. Real-time performance trending of cloud infrastructure (currently AWS).

• Kaavo's IMOD is an easy to use online application. Cloud Computing Made Easy.

• Microsoft Mesh

• Nasstar - SaaS provider. Business grade Hosted Desktop service, UK market leaders.

• Nirvanix - Cloud Storage

• TrustSaaS - uptime monitoring and alerting service ('SaaS Weather Report') for Software as a Service (SaaS) run by an independent third party.

Enomalism, convirt, redhat genome, hyperVM, lxlabs, LN, OpenNEbula, reservoir-fp7, scalr,eucalyptus,ganeti,gplhost，ovirt。

Useful open source projects to build cloud platform:

Kenso, hyperic, virt-P2V。

• 優(yōu)先訪問(wèn)權(quán)風(fēng)險(xiǎn)(Privileged user access.)

• 管理權(quán)限風(fēng)險(xiǎn)(Regulatory compliance.)

• 數(shù)據(jù)處所風(fēng)險(xiǎn)(Data location.)

• 數(shù)據(jù)隔離風(fēng)險(xiǎn)(Data segregation.)

• 數(shù)據(jù)恢復(fù)風(fēng)險(xiǎn)(Recovery.)

• 調(diào)查支持風(fēng)險(xiǎn)(Investigative support.)

• 長(zhǎng)期發(fā)展風(fēng)險(xiǎn)(Long-term viability.)

下面是Dell描述的云計(jì)算模型：

解釋：

這是一個(gè)明顯學(xué)術(shù)化的模型圖，就像ISO七層網(wǎng)絡(luò)模型，永遠(yuǎn)只有參考意義。

相比而言，下面這個(gè)示意圖更具有表現(xiàn)力：

下面是DELL描述的一個(gè)數(shù)據(jù)中心

事實(shí)上，這個(gè)數(shù)據(jù)中心和傳統(tǒng)的數(shù)據(jù)中心看起來(lái)沒(méi)有很大的不同，云計(jì)算的含義是不能通過(guò)一個(gè)數(shù)據(jù)中心示意圖來(lái)表達(dá)的。

市場(chǎng)分為如下幾個(gè)層次：

• Cloud Computing
• Platform as a Service (PaaS)
• Software as a Service (SaaS)
• Core Cloud Services

解釋如下：

• Infrastructure: the core computing resources and network fabric for the cloud deployment
• Platform: the software infrastructure that allows sys admins and developers to deploy an app to the cloud
• Core Services: additional services that can be woven into the cloud app, such as billing, storage, integration
• Applications: the ultimate cloud product - the actual cloud based application that the user touches. These number in the thousands.

市場(chǎng)視圖

下面是市場(chǎng)參與者列表：

• Public Cloud
  • Amazon EC2
  • ServePath GoGrid
  • Rackspace Mosso Cloud
  • Joyent Accelerators
  • AppNexus
  • Flexiscale
  • ElasticHosts
• Private Cloud
  • Eucalyptus
  • Cassatt Active Response
  • Enomaly Enomalism Platform
  • Grid
    • Globus Toolkit
    • Beowulf
    • Sun Project Caroline
• Open Cloud Platforms
  • Heroku
  • Morph Labs
  • Aptana CloudStudio
• Custom Cloud Platforms
  • Salesforce.com force.com
  • Google App Engine
  • Bungee Labs Connect
  • Intuit Quickbase
  • LongJump
  • Coghead
• Cloud Platform Tools
  • Fabric Mgmt
    • Rightscale
    • Scalr
    • Elastra Cloud Server
    • 3Tera AppLogic
    • Kaavo IMOD
  • Data Grids
    • Oracle Coherence
    • IBM eXtreme Scale
    • GigaSpaces Data Grid
    • Gemstone Gemfire
  • Virtual Appliances
    • rPath
    • CohesiveFT
  • Hyperic CloudStatus
  • Hadoop
• Storage
  • Amazon S3
  • Amazon SimpleDB
  • Microsoft SSDS
  • Rackspace Mosso CloudFS
  • Google BigTable
• Integration
  • Bungee Labs Connect
  • Boomi
  • MuleSource Mule OnDemand
  • Amazon SQS
  • Microsoft BizTalk Services
  • OpSource Connect
  • SnapLogic SaaS Solution Packs
  • gnip
  • CastIron
  • Appirio
  • Skemma
  • Appian Anywhere
• Value-Add
  • Billing
    • OpSource Billing
    • Aria
    • eVapt
    • Zuora
    • Vindicia
  • Security
    • Ping Identity
    • OpenID/OAuth
  • Strikeiron
• Applications
  • Salesforce.com
  • Netsuite
  • Taleo
  • Oracle OnDemand
  • Concur
  • Google Apps ^[18]