SlideShare a Scribd company logo
Cloud Computing: Where Do We Go
          from Here?


           Tzi-cker Chiueh (     )

     Cloud Computing Research Center for
         Mobile Applications (CCMA)



                  NCCU                                           1
                                     1
                                           Copyright 2008 ITRI
Cloud computing is hot!

But where is the $$ for Taiwan?



                                                2
             NCCU
                          Copyright 2008 ITRI
Cloud Computing Basics

• “ s cloud illusions I recall; I really don’know
   It’                                      t
  clouds at all”
     --- Both Sides Now, Joni Mitchell
• Our definition
   – Centralization of IT infrastructures
   – Usage-based resource allocation and charging
• Enabling technologies
   – HW/SW to build and run very large data centers
     efficiently
   – Scalable, reliable and secure cloud services
   – Faster Internet connectivity
                          NCCU                        3
Cloud Service Models

• Infrastructure as a Service (IaaS)
   – A set of virtual machines with storage space and external network
     bandwidth  unfurnished apartment
   – Example: Amazon Web Service
• Platform as a Service (PaaS)
   – An operating environment including (application-specific) libraries
     and supporting services (DBMS, AAA)  furnished apartment
   – Example: Google’ App Engine, Microsoft’ Azure, IBM’ XaaS
                      s                        s             s
• Software as a Service (SaaS)
   – Turn-key software hosted on the cloud and accessible through the
     browser  hotel
   – Example: salesforce.com, and all major desktop software vendors

                                 NCCU                                 4
Cloud Computing Food Chain

                                  Enterprise User
                    End User      Consumer
                                  SaaS User
                 SaaS
                                         Enterprise IT Department
            Cloud Service User           Web Application Developer
                                         SaaS Provider
            IaaS, PaaS

           Cloud Service Provider         Mega Data Center Operator
                                          Cloud Application Platform Provider
            HW                  SW

Mega Data Center               Mega Data Center
Hardware Provider              Software Provider
                                  NCCU                                   5
Taiwan’ Positioning in Cloud
           s
        Computing Food Chain
• Just end user?
• Sophisticated cloud service user or cloud application
  developer? Exportable?
• Operate its cloud service provider?
   – Data confidentiality concern (especially for E-government service)
• Serve as a major cloud service provider in a geographical
  region?
   – Sufficient bandwidth and connectivity?
   – Climate and geology factors are against us
• Become a major HW/SW solution provider for cloud
  service providers?
   – data center hardware component  total data center solution
   – ITRI/CCMA’ bet here
                 s
                                NCCU                                6
Microsoft’ Container-based Data
         s
       Center in Chicago




              NCCU                7
Inside the Container




         NCCU          8
Cloud-Scale Data Center

• > 10,000 servers  $50M USD apiece
• Technology components:
  – Container computer: Optimal HW building block for
    constructing a cloud data center
  – Cloud OS: An end-to-end software stack that runs
    cloud applications and operates a cloud data center
  – Non-ICT technology: seismic, fire, physical security, etc.
  – Integration/operation know-how: Operational
    experiences and expertise for putting together and
    running a cloud-scale data center


                           NCCU                            9
Data Center Solution Provider
• Cloud Hardware Platform
   –   Servers (HP/Dell)
   –   Storage arrays: primary and secondary (EMC/NetApp)
   –   Network switches: LAN and SAN (Cisco/Juniper/Brocade)
   –   Front-end Appliances:
         • IPS/Firewall/NAT (Checkpoint)
         • Traffic Shaper (Packeteer)
         • Server Load Balancer (F5)
         • Multi-Homing Load Balancer (BIG-IP)
• Cloud Software Platform
   –   Hypervisor (VMware, Citrix, Microsoft)
   –   Virtualization management (VMware, Citrix, Microsoft)
   –   Storage management (EMC, NetApp, HDS, IBM, Symantec)
   –   Backup/replication (Symantec)
   –   Data center automation + energy management (Tivoli, CA, BMC, HP)
   –   Security (Symantec, Cisco, McAfee)
• Non-ICT
   –   Power: Electricity distribution network (Delta/Emerson)
   –   Cooling: Thermal management (APC)
   –   Facilities: Fire, seismic, battery, physical security
   –   Regulations                          NCCU                          10
Data Center Design Trends
• Containerization
   – Best HW packaging for cloud data centers
      • Server  Server rack  container  warehouse
      • Commodity HW only, not a supercomputer
   – More efficient power distribution and thermal design
   – Unification of computing, memory, network and storage resources
      • Virtualization of all HW resources: Software-definable
        boundaries
   – Faster deployment: no on-premise installation needed
   – Requires light-out operation
• Google-style data center design
   – End-to-end redundancy
      • Failure is a common case

                                   NCCU                          11
Missing Links
• Goal: Enable Taiwan to build data centers for the world
• Container Computer 1.0
   – Manageable container computer
   – Differences between a set of servers/switches/storage boxes and a
     container computer?
       • Scalable storage/network architecture
       • System management software
       • Energy-efficient cooling
• Cloud Operating System 1.0
   – End-to-end data center software stack for supporting a AWS-like
     service on a set of commodity HW
   – Tight integration of storage, resource, security and system/network
     management
• First prototype of both expected at the end of 2010
                                   NCCU                              12
Container Computer 1.0

• Objective: Physical data center in a box
• Architecture Design Principles:
  – Commodity HW only
     • No storage box, appliance or accelerator
  – System-wide optimization
     • Component vs. self-contained system
     • server  container computer  warehouse computer
• Major features:
  – All-layer-2 data center network architecture
  – Scalable internet appliance logic
  – Touch cooling-based thermal management
                             NCCU                         13
Cloud Operating System 1.0
Objective: Multiplexing virtual data centers on a physical data center
    –   No. of VM instances each with CPU performance and memory size requirement
    –   Shared storage space requirement
    –   External network bandwidth requirement
    –   VPN setting
    –   Firewall/IDS/IPS rules
    –   Server load balancing policy
    –   Public IP addresses
    –   Network configuration, including DNS and DHCP
    –   OS and application image
Design Goals:
    –   Very very very very… easy to manage for both physical and virtual data
        center administrators
    –   Ability to hide HW failures from applications
    –   Intelligent data center-wide power management
    –   Multi-tenancy security
    –   Scalability across all workloads
                                      NCCU                                      14
Components in Cloud OS 1.0
                       Virtual Data Center Management Tool

   Primary/Secondary
  Storage Management        Virtual Data Center
                          Provisioning (VMware)         Network/System
         (EMC)                                           Management
                                                            (Tivoli)
Physical
Infrastructure            Physical Compute Servers
Management                                                     Security
                   Distributed Main/Secondary Storage
Tool (Dell)                                                  (Checkpoint)
                            All-layer-2 Network

                                                        Inter-VM Load
                                                        Balancing (F5)
     Power Management
           (Tivoli)       Inter-PM Load Balancing
                          and VM Fail-over (VMware)
                                    NCCU                                 15
Cloud OS 1.0 Summary
• World’ first integrated data center software stack that
       s
    – Integrates storage management, virtualization management,
      system/network management, security, and power management
    – Targets at cloud-scale data centers and supports virtual data center-
      based IaaS
• Competitions:
    – Virtual Computing Environment Coalition (Acadia): Cisco, EMC, and
      VMware
    – Microsoft/HP Cloud Alliance




                                    NCCU                                      16
Where CCMA Adds Value
SaaS    Applications       EMR     Smart Grid       Automated Cloudification

        Cloud Application Middleware Platform
PaaS      LAMP     .NET     WebSphere        WebLogic         Google App Engine


        Cloud System Software Platform
       Hypervisor   Virtualization Management Storage Management Security
       Backup/Replication     Data Center Automation Energy Management


IaaS     Cloud Hardware Platform
        Scalable System Architecture    System Management         Cooling



        Servers      Storage Arrays
                                 NCCU    Switches         Power Distribution
                                                                          17
Evolution Path
                     Taiwanese Companies’      Asia Pacific Data and
                     World-wide IT Center      Application Backup
Profit Margin
   Total Solution
   Provider




                          Cloud Data Center System Integrator
                                with Operator Know-how

                          Cloud Data Center System Integrator
   Plant
   Power




                       Total Container Computer Solution Provider
     Generator
     Power




                    Container Computer OEM Container Computer ODM

                         Container Computer Component ODM
                                        NCCU                        18
Success Criteria

• Bring up Taiwan’ computer HW industry to rise to
                 s
  the cloud computer era
  – A Taiwan-spec container computer that powers cloud-
    scale data centers in the world
• Jump-start Taiwan’ systems software industry
                   s
  – A Taiwan-based cloud OS company that provide end-to-
    end software stack solutions to cloud data centers in the
    world
• Container computer ODM  Red Hat  Huawei

                           NCCU                           19
15
                    /      (   240      )

1.                  FY99~FY103/37.2
2.                  FY99~FY103/1.25
3.                      FY99~FY103/10
4.                      FY99~FY103/15
5.                  FY99~FY103/6.98
6.                  FY100~FY103/65
7.                      FY100~FY103/4
8.                  FY99~FY103/16.5
9.                        FY100~103/6
10.                 FY100~FY103/6.1
11.                 FY99~ FY103/3.6
12.               FY100~FY102/13.17

13.                FY99~FY102/39.92

14.                  FY98~FY10 /7.5
15.               FY99~FY103 / 7.78
      NCCU                                  20
International Competition Landscape
• United States:
   – Holds dominant position in all cloud computing segments
   – No industrial development policy
• European Union
   – Leverages strength in telecom equipment/service business and focuses
     on “
        cloud service systems”for carriers
• Japan
   – Digital Japan Creation Project; Kasumigaseki cloud
   – NEC, Fujitsu, Hitachi enter the Cloud Infrastructure space
• Korea
   – National Computing and Information Agency : 40 agencies’   IT
     departments consolidated by December 2007
   – Carefully chosen cloud applications/services: IPTV, Green Internet Café,
     Mobile device, Smart Grid, On-line gaming, Tele-commuting, etc.
• China
   – Being formulated
   – Huawei has already grabbed most of China Mobile’ data center
                                                    s
     businesses
                                    NCCU                                   21
Conclusion

• Cloud computing is all about consolidation of IT
  infrastructures and usage-based resource allocation
• Most on-premise applications will eventually be cloudified
• Cloud-scale data center industry is emerging
• ITRI’ integrated data center solution, Container Computer
       s
  1.0 + Cloud OS 1.0, is expected to provide 80% of the
  functionalities at 1/2 cost of current market-leading
  solutions




                           NCCU                          22
Thank You!

Questions and Comments?

    tcc@itri.org.tw



          NCCU            23

More Related Content

ODP
Innovation for Participation - Paul De Decker, Sun Microsystems
PDF
Virtualization Approach: Theory and Application
PDF
HPC Cloud Computing with OpenNebula
PPTX
Cloud management
PPTX
Cloud and its job oppertunities
PDF
Cloud Computing_2015_03_05
PDF
Cloud Computing System models for Distributed and cloud computing & Performan...
PPT
IBM SmartCloud Orchestration
Innovation for Participation - Paul De Decker, Sun Microsystems
Virtualization Approach: Theory and Application
HPC Cloud Computing with OpenNebula
Cloud management
Cloud and its job oppertunities
Cloud Computing_2015_03_05
Cloud Computing System models for Distributed and cloud computing & Performan...
IBM SmartCloud Orchestration

What's hot (19)

PDF
Innovation in cloud computing architectures with open nebula
PPT
Cc unit 2 ppt
PDF
Presentation introduction to cloud computing and technical issues
PDF
Cloud & Data Center Networking
PPT
Lessons Learned during IBM SmartCloud Orchestrator Deployment at a Large Tel...
PDF
Реализация частной и гибридной облачной IT-инфраструктуры предприятия на осно...
DOCX
Cloud computing
PDF
Основы построения облачной IT-инфраструктуры. Особенности и преимущества суще...
PPTX
Multi-Tenancy and Virtualization in Cloud Computing
PPTX
Data Tactics dhs introduction to cloud technologies wtc
PPTX
Cloud Delivery Model Considerations
PPT
IBM Private Modular Cloud
PDF
Cloud security deep dive infoworld jan 2011
PPT
IBM Cloud Journey v10
PDF
Virtualization - cloud computing
PPT
UTF-8'en'IBM_Cloud_SCO_Content_20130702c
PDF
Week 1 lecture material cc
PDF
The unified data center for cloud david yen
DOCX
Quiz 1 cloud computing
Innovation in cloud computing architectures with open nebula
Cc unit 2 ppt
Presentation introduction to cloud computing and technical issues
Cloud & Data Center Networking
Lessons Learned during IBM SmartCloud Orchestrator Deployment at a Large Tel...
Реализация частной и гибридной облачной IT-инфраструктуры предприятия на осно...
Cloud computing
Основы построения облачной IT-инфраструктуры. Особенности и преимущества суще...
Multi-Tenancy and Virtualization in Cloud Computing
Data Tactics dhs introduction to cloud technologies wtc
Cloud Delivery Model Considerations
IBM Private Modular Cloud
Cloud security deep dive infoworld jan 2011
IBM Cloud Journey v10
Virtualization - cloud computing
UTF-8'en'IBM_Cloud_SCO_Content_20130702c
Week 1 lecture material cc
The unified data center for cloud david yen
Quiz 1 cloud computing
Ad

Similar to 2010 09-24-闕志克老師-cloud computing where do we go (20)

PPT
PPTX
Cloud computing and its job opportunities
PPTX
Introduction to Cloud Computing
PDF
9-cloud-computing.pdf
PPTX
Cloud computing by prabhunath sharma
PPT
Fundamentals of Cloud Computing Basics.ppt
PPT
Cloud computing
PPTX
Cloud Computing
DOCX
Cloud Computing presentation . docx
PPTX
Introduction to Cloud seminar presentation.pptx
PDF
Cloud white paper v3.0
PPTX
Cloud virtualization
PPTX
Clould Computing and its application in Libraries
PDF
Cloud computing
PPTX
cloude computing
PDF
OIT552 Cloud Computing Material
PDF
CC Notes.pdf of jdjejwiwu22u28938ehdh3y2u2838e
PPT
Salesforce - classification of cloud computing
PPTX
Gef 2012 InduSoft Presentation
PPT
Introduction to Cloud Computing
Cloud computing and its job opportunities
Introduction to Cloud Computing
9-cloud-computing.pdf
Cloud computing by prabhunath sharma
Fundamentals of Cloud Computing Basics.ppt
Cloud computing
Cloud Computing
Cloud Computing presentation . docx
Introduction to Cloud seminar presentation.pptx
Cloud white paper v3.0
Cloud virtualization
Clould Computing and its application in Libraries
Cloud computing
cloude computing
OIT552 Cloud Computing Material
CC Notes.pdf of jdjejwiwu22u28938ehdh3y2u2838e
Salesforce - classification of cloud computing
Gef 2012 InduSoft Presentation
Introduction to Cloud Computing
Ad

More from nccuscience (10)

PDF
2010 09-20-張志威老師-機器學習到人類創新
PDF
2010 09-17-張志威老師-confucius & “its” intelligent disciples
PDF
2010 07-26-interdisciplinary research and learning
PPT
Science talk 100510-李湘楠
PPT
Science talk 100315-蔡介立
PPT
Science Talk-091221-劉昭麟
PDF
Science Talk-091123-廖瑞銘
PPT
Science Talk-091102-郭光宇
PPT
Science Talk-100111-陸行
PDF
Science Talk-091012-楊健銘
2010 09-20-張志威老師-機器學習到人類創新
2010 09-17-張志威老師-confucius & “its” intelligent disciples
2010 07-26-interdisciplinary research and learning
Science talk 100510-李湘楠
Science talk 100315-蔡介立
Science Talk-091221-劉昭麟
Science Talk-091123-廖瑞銘
Science Talk-091102-郭光宇
Science Talk-100111-陸行
Science Talk-091012-楊健銘

Recently uploaded (20)

PDF
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
PDF
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
DOCX
The AUB Centre for AI in Media Proposal.docx
PDF
Chapter 3 Spatial Domain Image Processing.pdf
PDF
Per capita expenditure prediction using model stacking based on satellite ima...
PDF
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
PPTX
Spectroscopy.pptx food analysis technology
PDF
Review of recent advances in non-invasive hemoglobin estimation
PDF
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
PDF
Machine learning based COVID-19 study performance prediction
PDF
Unlocking AI with Model Context Protocol (MCP)
PPTX
Digital-Transformation-Roadmap-for-Companies.pptx
PDF
cuic standard and advanced reporting.pdf
PPTX
MYSQL Presentation for SQL database connectivity
PDF
Encapsulation theory and applications.pdf
PDF
Diabetes mellitus diagnosis method based random forest with bat algorithm
PDF
How UI/UX Design Impacts User Retention in Mobile Apps.pdf
PPTX
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
PDF
Empathic Computing: Creating Shared Understanding
PPT
Teaching material agriculture food technology
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
The AUB Centre for AI in Media Proposal.docx
Chapter 3 Spatial Domain Image Processing.pdf
Per capita expenditure prediction using model stacking based on satellite ima...
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
Spectroscopy.pptx food analysis technology
Review of recent advances in non-invasive hemoglobin estimation
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
Machine learning based COVID-19 study performance prediction
Unlocking AI with Model Context Protocol (MCP)
Digital-Transformation-Roadmap-for-Companies.pptx
cuic standard and advanced reporting.pdf
MYSQL Presentation for SQL database connectivity
Encapsulation theory and applications.pdf
Diabetes mellitus diagnosis method based random forest with bat algorithm
How UI/UX Design Impacts User Retention in Mobile Apps.pdf
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
Empathic Computing: Creating Shared Understanding
Teaching material agriculture food technology

2010 09-24-闕志克老師-cloud computing where do we go

  • 1. Cloud Computing: Where Do We Go from Here? Tzi-cker Chiueh ( ) Cloud Computing Research Center for Mobile Applications (CCMA) NCCU 1 1 Copyright 2008 ITRI
  • 2. Cloud computing is hot! But where is the $$ for Taiwan? 2 NCCU Copyright 2008 ITRI
  • 3. Cloud Computing Basics • “ s cloud illusions I recall; I really don’know It’ t clouds at all” --- Both Sides Now, Joni Mitchell • Our definition – Centralization of IT infrastructures – Usage-based resource allocation and charging • Enabling technologies – HW/SW to build and run very large data centers efficiently – Scalable, reliable and secure cloud services – Faster Internet connectivity NCCU 3
  • 4. Cloud Service Models • Infrastructure as a Service (IaaS) – A set of virtual machines with storage space and external network bandwidth  unfurnished apartment – Example: Amazon Web Service • Platform as a Service (PaaS) – An operating environment including (application-specific) libraries and supporting services (DBMS, AAA)  furnished apartment – Example: Google’ App Engine, Microsoft’ Azure, IBM’ XaaS s s s • Software as a Service (SaaS) – Turn-key software hosted on the cloud and accessible through the browser  hotel – Example: salesforce.com, and all major desktop software vendors NCCU 4
  • 5. Cloud Computing Food Chain Enterprise User End User Consumer SaaS User SaaS Enterprise IT Department Cloud Service User Web Application Developer SaaS Provider IaaS, PaaS Cloud Service Provider Mega Data Center Operator Cloud Application Platform Provider HW SW Mega Data Center Mega Data Center Hardware Provider Software Provider NCCU 5
  • 6. Taiwan’ Positioning in Cloud s Computing Food Chain • Just end user? • Sophisticated cloud service user or cloud application developer? Exportable? • Operate its cloud service provider? – Data confidentiality concern (especially for E-government service) • Serve as a major cloud service provider in a geographical region? – Sufficient bandwidth and connectivity? – Climate and geology factors are against us • Become a major HW/SW solution provider for cloud service providers? – data center hardware component  total data center solution – ITRI/CCMA’ bet here s NCCU 6
  • 7. Microsoft’ Container-based Data s Center in Chicago NCCU 7
  • 9. Cloud-Scale Data Center • > 10,000 servers  $50M USD apiece • Technology components: – Container computer: Optimal HW building block for constructing a cloud data center – Cloud OS: An end-to-end software stack that runs cloud applications and operates a cloud data center – Non-ICT technology: seismic, fire, physical security, etc. – Integration/operation know-how: Operational experiences and expertise for putting together and running a cloud-scale data center NCCU 9
  • 10. Data Center Solution Provider • Cloud Hardware Platform – Servers (HP/Dell) – Storage arrays: primary and secondary (EMC/NetApp) – Network switches: LAN and SAN (Cisco/Juniper/Brocade) – Front-end Appliances: • IPS/Firewall/NAT (Checkpoint) • Traffic Shaper (Packeteer) • Server Load Balancer (F5) • Multi-Homing Load Balancer (BIG-IP) • Cloud Software Platform – Hypervisor (VMware, Citrix, Microsoft) – Virtualization management (VMware, Citrix, Microsoft) – Storage management (EMC, NetApp, HDS, IBM, Symantec) – Backup/replication (Symantec) – Data center automation + energy management (Tivoli, CA, BMC, HP) – Security (Symantec, Cisco, McAfee) • Non-ICT – Power: Electricity distribution network (Delta/Emerson) – Cooling: Thermal management (APC) – Facilities: Fire, seismic, battery, physical security – Regulations NCCU 10
  • 11. Data Center Design Trends • Containerization – Best HW packaging for cloud data centers • Server  Server rack  container  warehouse • Commodity HW only, not a supercomputer – More efficient power distribution and thermal design – Unification of computing, memory, network and storage resources • Virtualization of all HW resources: Software-definable boundaries – Faster deployment: no on-premise installation needed – Requires light-out operation • Google-style data center design – End-to-end redundancy • Failure is a common case NCCU 11
  • 12. Missing Links • Goal: Enable Taiwan to build data centers for the world • Container Computer 1.0 – Manageable container computer – Differences between a set of servers/switches/storage boxes and a container computer? • Scalable storage/network architecture • System management software • Energy-efficient cooling • Cloud Operating System 1.0 – End-to-end data center software stack for supporting a AWS-like service on a set of commodity HW – Tight integration of storage, resource, security and system/network management • First prototype of both expected at the end of 2010 NCCU 12
  • 13. Container Computer 1.0 • Objective: Physical data center in a box • Architecture Design Principles: – Commodity HW only • No storage box, appliance or accelerator – System-wide optimization • Component vs. self-contained system • server  container computer  warehouse computer • Major features: – All-layer-2 data center network architecture – Scalable internet appliance logic – Touch cooling-based thermal management NCCU 13
  • 14. Cloud Operating System 1.0 Objective: Multiplexing virtual data centers on a physical data center – No. of VM instances each with CPU performance and memory size requirement – Shared storage space requirement – External network bandwidth requirement – VPN setting – Firewall/IDS/IPS rules – Server load balancing policy – Public IP addresses – Network configuration, including DNS and DHCP – OS and application image Design Goals: – Very very very very… easy to manage for both physical and virtual data center administrators – Ability to hide HW failures from applications – Intelligent data center-wide power management – Multi-tenancy security – Scalability across all workloads NCCU 14
  • 15. Components in Cloud OS 1.0 Virtual Data Center Management Tool Primary/Secondary Storage Management Virtual Data Center Provisioning (VMware) Network/System (EMC) Management (Tivoli) Physical Infrastructure Physical Compute Servers Management Security Distributed Main/Secondary Storage Tool (Dell) (Checkpoint) All-layer-2 Network Inter-VM Load Balancing (F5) Power Management (Tivoli) Inter-PM Load Balancing and VM Fail-over (VMware) NCCU 15
  • 16. Cloud OS 1.0 Summary • World’ first integrated data center software stack that s – Integrates storage management, virtualization management, system/network management, security, and power management – Targets at cloud-scale data centers and supports virtual data center- based IaaS • Competitions: – Virtual Computing Environment Coalition (Acadia): Cisco, EMC, and VMware – Microsoft/HP Cloud Alliance NCCU 16
  • 17. Where CCMA Adds Value SaaS Applications EMR Smart Grid Automated Cloudification Cloud Application Middleware Platform PaaS LAMP .NET WebSphere WebLogic Google App Engine Cloud System Software Platform Hypervisor Virtualization Management Storage Management Security Backup/Replication Data Center Automation Energy Management IaaS Cloud Hardware Platform Scalable System Architecture System Management Cooling Servers Storage Arrays NCCU Switches Power Distribution 17
  • 18. Evolution Path Taiwanese Companies’ Asia Pacific Data and World-wide IT Center Application Backup Profit Margin Total Solution Provider Cloud Data Center System Integrator with Operator Know-how Cloud Data Center System Integrator Plant Power Total Container Computer Solution Provider Generator Power Container Computer OEM Container Computer ODM Container Computer Component ODM NCCU 18
  • 19. Success Criteria • Bring up Taiwan’ computer HW industry to rise to s the cloud computer era – A Taiwan-spec container computer that powers cloud- scale data centers in the world • Jump-start Taiwan’ systems software industry s – A Taiwan-based cloud OS company that provide end-to- end software stack solutions to cloud data centers in the world • Container computer ODM  Red Hat  Huawei NCCU 19
  • 20. 15 / ( 240 ) 1. FY99~FY103/37.2 2. FY99~FY103/1.25 3. FY99~FY103/10 4. FY99~FY103/15 5. FY99~FY103/6.98 6. FY100~FY103/65 7. FY100~FY103/4 8. FY99~FY103/16.5 9. FY100~103/6 10. FY100~FY103/6.1 11. FY99~ FY103/3.6 12. FY100~FY102/13.17 13. FY99~FY102/39.92 14. FY98~FY10 /7.5 15. FY99~FY103 / 7.78 NCCU 20
  • 21. International Competition Landscape • United States: – Holds dominant position in all cloud computing segments – No industrial development policy • European Union – Leverages strength in telecom equipment/service business and focuses on “ cloud service systems”for carriers • Japan – Digital Japan Creation Project; Kasumigaseki cloud – NEC, Fujitsu, Hitachi enter the Cloud Infrastructure space • Korea – National Computing and Information Agency : 40 agencies’ IT departments consolidated by December 2007 – Carefully chosen cloud applications/services: IPTV, Green Internet Café, Mobile device, Smart Grid, On-line gaming, Tele-commuting, etc. • China – Being formulated – Huawei has already grabbed most of China Mobile’ data center s businesses NCCU 21
  • 22. Conclusion • Cloud computing is all about consolidation of IT infrastructures and usage-based resource allocation • Most on-premise applications will eventually be cloudified • Cloud-scale data center industry is emerging • ITRI’ integrated data center solution, Container Computer s 1.0 + Cloud OS 1.0, is expected to provide 80% of the functionalities at 1/2 cost of current market-leading solutions NCCU 22
  • 23. Thank You! Questions and Comments? tcc@itri.org.tw NCCU 23