On Cloud Computing: The Dominance and Governance on The Computing Landscape By Shang-Sheng Jeng March 22 2010Slide 2
Baseline Consideration: A New Dimension: Cyberspace A New Identity: Cybog A New Relationship and AssetSlide 3
Baseline Consideration: Information Intelligence Instruction This is a Competitive World From Profitability to Survivability Computing Paradigm Shift The Next Thirty Years of WorldSlide 4
"The principal need article to decimate for tanks is not the adversary but rather the counter tank weapons." " The will to survival"Slide 5
Presentation Objectives: Offensive: To Manipulate the "Distributed computing" Landscape Opportunity & Capability To Balance the "Distributed computing" Competition IBM (Oracle+Sun), Google (MS+Yahoo) To Dominate the Government Cloud in Taiwan and China To Lead the Reform of "Programming Industry"Slide 6
Presentation Objectives: Defensive: Make Taiwan\'s Computer Industry make due in Computing Paradigm Shift Take the monetary and mechanical advantage for Taiwan\'s Cloud Computing Users (Institutional or Personal) Minimize the lost of Taiwan\'s product industry Find a Way to Keep Taiwan\'s Software Industry SurviveSlide 7
Content 1. Inceptions 2. Players 3. Patterns 4. ReactionSlide 8
1. Beginnings 1.1 Industry Restructuring 1.2 Technology Maturity 1.3 Issue, Innovation and CompetitionSlide 9
1.1 Industry Restructuring (old, vertical) Network CISCO Server Sun Micro, HP DB Oracle Services IBM Business SAPSlide 10
1.1 Industry Restructuring (now, coordination) Market Overlapping Market Saturation Requirement Dynamics and Technology Complexity pushed Outsourcing (for clients) and M&A (for suppliers)Slide 11
1.2 Technology Maturity Producer Processing Communication Storage Consumer Web based Services Architecture Clustering, Grid, Data CenterSlide 12
1.3 Issue, Innovation, and Competition Open Source Movement Software as a Service Hardware, Infrastructure, Platform, as a Service Web 2.0 Push Concept of Utility Computing (Nicholas Carr, "IT doesn\'t Make a difference", "Huge Switch")Slide 13
1.3 Issue, Innovation, and Competition Data as Asset Biz Model Prosumer (maker as shopper and the other way around) Biz Model SOA & Event-driven Computing (C4KISR, SOA-R, Cloud Computing) Functional Cloud Integrated Service (illustration: Chiang\'s remembrance pages)Slide 14
Web 2.0 Push The Characteristics of Web 2.0 Virtual Disk Xdriver, iDisk Note BuzzWord Blogger WordPress FeedReader Rojo, Bloglines Photo Flickr, Photobucket Video YouTube, JoostSlide 15
Web 2.0 Push eMail gmail. Hotmail iPhone skype Social network MySpace, Facebook Virtual ID Second Life, World of WarcraftSlide 16
2. Player 2.1 Type 1 Service: Serve to Government and Large Institutions 2.2 Type 2 Service: Serve to Mass Individual User, and Small Business 2.3 Type 3 Service: Serve for Special PurposeSlide 17
2.1 Type 1 Player Oriented IBM* (Key Player) Oracle EMC HP SAPSlide 18
IBM Outsourcing Service: Applications Outsourcing, Business Process Outsourcing, IT Outsourcing and Hosting Service Science Mainframe Computer and Services Integrated Datacenter (GIE)Slide 19
IBM Blue Cloud Almaden Research Center: Xen, PowerVM, Linux, Hadoop File System (Open Source) Tivoli, DB2, WebSphere, x86 Blade (Cash Cow)Slide 20
Services Sciences Customer Insights Social Analytics Services Software Engineering Systems Management and Service Management Social SimulationSlide 21
Service Science Service Science, Management, and Engineering IBM System Journal, Volume 47, Number 1, 2008Slide 22
IBM Penetration IBM Software Available Pay-As-You-Go on EC2 Monday, February 23, 2009Slide 23
Oracle Business OnLine Siebel Sun Micro: Java, Virtual On-interest Data Center, BlackboxSlide 24
EMC VMware, Pi, Smarts, Berkeley Data Systems, MozySlide 25
2.2 Type 2 Player Oriented Amazon Google* (Key Player) Salesforce Microsoft YahooSlide 26
Amazon Redundant Resources Services as a control apparatus Amazon Web Services (AWS): Simple Storage Service (S3), Elastic Compute Cloud (EC2), Simple Queuing Service, SimpleDBSlide 27
Google Enterprise Cloud and Consumer Cloud Google App Engine Google File System, Map/Reduce, Chubby, BigTable 。Slide 28
Salesforce Software as a Service, Platform as a Service AppExchange: Customer Relationship Management (CRM), Enterprise Content Management (ECM) First abundance 100M organization of just cloud sort administrationsSlide 29
Microsoft Dynamics CRM Online, Exchange Online, Office Communications Online Windows Live, Office Live, Xbox LiveSlide 30
Yahoo Hadoop Core (Open Source): a subproject of ApacheSlide 31
2.3 Type 3 Player Oriented Trend Micro Panda Cloud Antivirus PRCSlide 32
Trend Micro Security Cloud Smart Protection Network, In-the-Cloud administrationSlide 33
Panda Cloud Antivirus Light, Secure, Ease CloudAntivirus.exe (19,014,048 byte, May 10 2009) it have to uninstall Norton 360.Slide 34
3. Patterns 3.1 What is Cloud Computing 3.2 For Type 1 Oriented Cloud Computing 3.3 For Type 2 Oriented Cloud Computing 3.4 For Type 3 Oriented Cloud Computing 3.5 McKinsey\'s Recent Findings (April 2009) 3.6 Forrester\'s Opinions (April 13 2009) 3.7 US Government IT SpendingSlide 35
Trends 3.8 The US Cyber ChallengeSlide 36
3.1 What is Cloud Computing Defined by Reliable, Adaptable, Distributed Systems Lab , UC Berkeley (RAID, RISC) (support: Sun, Google, Microsoft) NIST Open Cloud ManifestoSlide 37
What is Cloud Computing? A Berkeley View of Cloud Computing (Feb 2009) Old thought: Software as a Service (SaaS) Def: conveying applications over the Internet Recently: "[Hardware, Infrastrucuture, Platform] as an administration" Poorly characterized so we stay away from all "X as an administration"Slide 38
What is Cloud Computing? (Berkeley) Utility Computing: pay-as-you-go figuring Illusion of unending assets No in advance cost Fine-grained charging (e.g. hourly)Slide 39
Spectrum of Clouds (Berkeley) Lower-level, Less administration Higher-level, More administration EC2 Azure AppEngine Force.com 39 Instruction Set VM (Amazon EC2, 3Tera) Bytecode VM (Microsoft Azure) Framework VM Google AppEngine, Force.comSlide 40
A Working Definition of Cloud Computing (NIST) Cloud processing is a compensation for every utilization model for empowering helpful, on-interest system access to a mutual pool of configurable and solid figuring assets (e.g., systems, servers, stockpiling, applications, benefits) that can be quickly provisioned and discharged with insignificant shopper administration exertion or administration supplier connection.Slide 41
A Working Definition of Cloud Computing (NIST) The cloud model is involved five key qualities, three conveyance models, and four organization models.Slide 42
5 Key Cloud Characteristics On-interest self-administration Ubiquitous system access Resource pooling Rapid flexibility Pay per useSlide 43
3 Cloud Delivery Models Cloud Software as a Service ( SaaS ) –Use supplier\'s applications over a system Cloud Platform as a Service ( PaaS ) Deploy client made applications to a Cloud Infrastructure as a Service ( IaaS ) Rent preparing, capacity, system limit, and other crucial registering assetsSlide 44
3 Cloud Delivery Models To be considered "cloud" they should be sent on top of cloud base that empowers the key qualitiesSlide 45
4 Cloud Deployment Models Private cloud undertaking claimed or rented Community cloud shared base for particular group Public cloud Sold to people in general, uber scale framework Hybrid cloud structure of two or more mistsSlide 46
4 Cloud Deployment Models Two sorts: inward and outside (my sentiment)Slide 47
Common Cloud Characteristics (NIST) Cloud figuring regularly influences: Massive scale Virtualization Free programming Autonomic processing Multi-tenure Geographically dispersed frameworks Advanced security innovationsSlide 48
What is Cloud Computing The Open Cloud Manifesto The key attributes of the cloud are the capacity to scale and arrangement figuring power powerfully in a cost effective manner and the capacity of the customer (end client, association or IT staff) to take advantage of that force without managing the fundamental multifaceted nature of the innovation. The cloud design itself can be private (facilitated inside an association\'s firewall) or open (facilitated on the Internet).Slide 49
The Goals of an Open Cloud Choice Flexibility Speed and Agility SkillsSlide 50
Choice (antilock-in) As an association picks a supplier or design or use show, an open cloud will make it simple for them to utilize an alternate supplier or engineering as the business 5 environment changes. In the event that the association needs to change suppliers due to new organizations, obtaining, client solicitations or government controls, they can do as such effectively. On the off chance that the association sends a private cloud, they can pick between suppliers as they amplify their ability and/or practical capacities. Assets that would have been spent on a troublesome relocation can rather be spent on development.Slide 51
Flexibility (bury correspondence and operation) No matter which cloud supplier and design an association uses, an open cloud makes it simple for them to work with different gatherings, regardless of the possibility that those different gatherings pick diverse suppliers and models. An open cloud will make it simple for associations to interoperate between various cloud suppliers.Slide 52
Speed and Agility (open interface) One of the worth suggestions of distributed computing is the capacity to scale equipment and programming as required. Utilizing open interfaces permits associations to fabricate new arrangements that coordinate open cloud
Principle danger of worry to optometry is vCJD as it requires the most ... GOS terms of administ ...
Measuring Natural Administration. Prologue to the Nearby Administration Gauge. Agreement DG/NRM ...
OECD, PUMA Policy Brief No 4. Administration and Anticorruption Core Course 24. The World Bank . ...
Plot. What is it?Why now?Cloud executioner appsEconomics for usersEconomics for providersChallen ...
Barters. Barters are generally utilized by organizations, private people and government offices ...
Cloud Forming Processes. A cloud with a volume of 1 cubic km contains around a million liters of ...
INTRODUCTION TO RESEARCH GOVERNANCE. The plan for today. Why Research Governance Research ...
Windows Azure for Research Roger Barga & Jared Jackson Contributors include Nelson Araujo, De ...
The Cloud Tutorial. Dan Reed, Roger Barga , Dennis Gannon Microsoft Research eXtreme Com ...
Windows Azure Building web sites and services in the cloud. Manu Cohen- Yashar Sela Group ...
Nov.8, 2010. Kai Hwang, USC. 2. Convenient Tools We Use over the Evolutional Periods In History ...
Diagram. In this unit, we investigate the idea of predominant strategiesDominance regularly requ ...
. Bring issues to light of hand strength and talk about treatment choices for re-preparing/excha ...
IntroductionIT LandscapeTop PrioritiesCloud DefinitionBasic ConceptsAppeal of CloudCommon Models ...
Diagnostics: Single Server versus the Cloud. Single Server. Static EnvironmentSingle surely unde ...
2. Cloud Agenda. Geoffrey Fox (Indiana U.) Remarks on Cloud ComputingMartin Swany (Internet2) Cl ...
Popularity based Governance-Good Governance. Standards. Popular government. Standard of Law. Fun ...