SlideShare une entreprise Scribd logo
1  sur  32
Social Cloud Computing
Simon Caton                                              http://www.facebook.com/SocialCloudComputing
                                                                       http://www.ksri.kit.edu/SocialCloud


KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




KIT – University of the State of Baden-Württemberg and
National Research Center of the Helmholtz Association                                             www.kit.edu
7 years of Cloud with and the same old hurdles

 Security
 Lack of Customisability
 Economics
    Small scale consumers have ad
     hoc requirements
    Providers have explicit
     incentives to lock in consumers
    Countless attempts are yet to
     produce the open cloud market
 Trust
    always assumed at some level
    Anonymity (Market-based/broker
     allocation)
    Many models fall apart when this
     is removed

                                        Karlsruhe Service Research Institute
                                                           www.ksri.kit.edu
Collaborative (Computing) Environments

 Users are lost in layer upon layer of abstraction
 Often make abhorrent trust assumptions
    Certificates!: Represent little more than an underlying social
     relationship in a dehumanized format
 Massively specified – but limited in capability
    Cannot see beyond the defined horizon
    Usually single purpose (at most few purposes)


 Examples:




                                                  Karlsruhe Service Research Institute
                                                                     www.ksri.kit.edu
Social Networks

 Ubiquitous: Facebook surpassed 1 billion users
 Represent mostly pre-existing real world relationships
 Have notions of pre-existent trust fabric inherently
  interwoven into the network structure
 Many applications now use social networks as a platform
  for:
    Authentication e.g. Facebook Connect
    Online Presence e.g. fb.com/your_page, Google Places
    Application Portals e.g. progress thru processors, ASPEN and
     PolarGrid project




                                               Karlsruhe Service Research Institute
                                                                  www.ksri.kit.edu
A Social Cloud
                                  Resources are idle 40-95%
          1,000,000,000 Users




        On average 190 friends
                                 Users contribute to “good” causes




   Social Cloud: a resource, service and capability sharing
    framework utilizing relationships established between
                members of a social network
                                              Karlsruhe Service Research Institute
                                                                 www.ksri.kit.edu
Talk Overview


     Vision of a Social Cloud


         Feasibility Study of a Social Market


          Platform Design


          Interaction: as a series of social and cognitive processes


         Use Cases: a Social CDN for academics, a social volunteer cloud


     Summary and Current Work


6                                                 Karlsruhe Service Research Institute
                                                                     www.ksri.kit.edu
Social Cloud Vision




                    Social
                  Exchange
                  Platform



                      Social Cloud
                                     Karlsruhe Service Research Institute
                                                        www.ksri.kit.edu
Architecture: High Level




   Shared Access
   Ownership
   Social Tie


   “Resource”


   Infrastructure
   Server
   Social
   Middleware

                           Karlsruhe Service Research Institute
                                              www.ksri.kit.edu
Is a Social Cloud Feasible?
    From: Chard, Caton, Rana and Bubendorfer; Social Cloud: Cloud Computing in
    Social Networks; IEEE Cloud 2010


    KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




9   KIT – University of the State of Baden-Württemberg and
    National Research Center of the Helmholtz Association             www.kit.edu
Architecture: Proof-of-concept Implementation




                            Agreement




10                                      Karlsruhe Service Research Institute
                                                           www.ksri.kit.edu
Feasibility Study: IEEE Cloud 2010

 Can a Social Cloud Scale?
 What are the computational requirements for an avg. SN?
 Can a Social Cloud function in a timely manner as a
  Facebook application?




                 Run on a single desktop
11                                     Karlsruhe Service Research Institute
                                                          www.ksri.kit.edu
Constructing a Social Cloud Platform
     From: Haas, Caton, Chard and Weinhardt; Co-Operative Infrastructures: An
     Economic Model for Providing Infrastructures for Social Cloud Computing;
     HICSS 2013

     KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




12   KIT – University of the State of Baden-Württemberg and
     National Research Center of the Helmholtz Association             www.kit.edu
How do we build a platform for a Social Cloud?




                   Platform needs
                   resources to:
                      Calculate allocations
                      Save bids and asks
                      Answer queries
                      …



13                                             Karlsruhe Service Research Institute
                                                                  www.ksri.kit.edu
Platform Design via Co-op Infrastructures

        A co-op is a scalable computing platform where all
       (computational) resources constituting the platform's
     infrastructure, as well as those made available over the
        platform, are owned and/or managed by its users.




                          What contribution schemes can secure the
                          resources are needed to keep a platform
                                         accessible?
14                                            Karlsruhe Service Research Institute
                                                                 www.ksri.kit.edu
Setting

 Platform Load: interpolated version of Cloud paper for a
  range of users [10, 400]

 Resource availability via SETI@home user distributions
 Users are modeled with varied compute resources
 Put these together and we “know” the total contribution
  needed

 Load is almost worst case:
      every trade occurs nearly in parallel
      half the social cloud takes part (if they meet min requirements)
      we include simple levels of redundancy

15                                                  Karlsruhe Service Research Institute
                                                                       www.ksri.kit.edu
Contribution Percentage
                                0.9


     Availability Percentage
                                                                                                                              0.4
                                0.8                                                                                          0.35
                                0.7                                                                                           0.3
                                0.6
Fixed Contribution              0.5
                                0.4
                                                                                                                             0.25
                                                                                                                              0.2
                                0.3                                                                                          0.15
                                0.2                                                                                           0.1
                                 Idea: Users have to provide a given percentage of their available resources to
                                0.1                                                                                          0.05
                                  0                                                                                             0
                                  the co-operative infrastructure 400
                                      10   20     50       100   200            10     20    50       100   200 400
                                 Questions:     Number of Users                            Number of Users

                                      rho_star do we set the percentage?
                                           How rho_star*1.1 rho_star*1.2 rho_star*1.5 rho_star rho_star*1.1 rho_star*1.2 rho_star*1.5

a)                                    System the effect of the contribution percentage on system reliability? of users, worst
                                           What is availability, worst case b) Average contribution
requirements                                                                                      case requirements
                                  1                                                                                          0.45




                                                                                                   Contribution Percentage
                                0.9
      Availability Percentage




                                                                                                                              0.4
                                0.8                                                                                          0.35
                                0.7
                                                                                                                              0.3
                                0.6
                                0.5                                                                                          0.25
                                0.4                                                                                           0.2
                                0.3                                                                                          0.15
                                0.2                                                                                           0.1
                                0.1                                                                                          0.05
                                  0                                                                                             0
                                          10     20         50       100         200      400                                            10     20        50       100           200          400
                                                           Number of Users                                                                               Number of Users
                                      rho_star   rho_star*1.1     rho_star*1.2     rho_star*1.5
                                                                                                                                    rho_star   rho_star*1.1      rho_star*1.2         rho_star*1.5

d)    System     availability,                                                         average    e) Average contribution of                                                                users,
         Platform Availability                                                                              Av. % Contribution
requirements                                                                                      average requirements
16                                                                                                                                             Karlsruhe Service Research Institute
                                                                                                                                                                  www.ksri.kit.edu
Voluntary Contribution
    Idea: Let users choose the amount of resources they contribute, based on
     individual preferences
    Approach:
       User behavior modeled through Utility Functions with Other-Regarding Preferences
        (User Types: self-interested, altruist, hybrid)
       Study dependence of system performance on the distribution of user types
       Price variable to capture relative ease/difficulty to provide resources




             Platform Availability                         Av. % Contribution

17                                                           Karlsruhe Service Research Institute
                                                                                www.ksri.kit.edu
Interaction in a Social Cloud as Social and Cognitive
     Processes
     From: Caton, Dukat, Grenz, Haas, Pfadenhauer and Weinhardt; Foundations of
     Trust: Contextualising Trust in Social Clouds; IEEE Social Computing and Its
     Applications 2012

     KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




18   KIT – University of the State of Baden-Württemberg and
     National Research Center of the Helmholtz Association                www.kit.edu
Interaction: Social and Cognitive Processes


                                 Social Cloud
  Ex-Ante                                     Social                                         Ex-Post
• Motivation: Outcome, Social
  Context, History
                                           Interchange                   • Feedback: Locally and to
                                                                           Network(s)
• Excess driven demand          • Formal Processes:                      • Recommendations: Rewards
• Demand induced Social           Initialisation, Identification,          and Sanctions
  Capital                         Allocation, Provisioning               • Interaction Archiving
                                • Informal Social                          (History)
                                  Communication and
                                  Coordination
           Prior
                                                                                        Completion
        Expectations                    Evolution of Relationship(s)




  19                                                                Karlsruhe Service Research Institute
                                                                                       www.ksri.kit.edu
 … an attribute of a relationship
 … in the necessary competence to be able to deliver
 … to deliver, keep promises etc.
20       Participant Selection for an Experimental Social Cloud   Karlsruhe Service Research Institute
                                                                                     www.ksri.kit.edu
What characterizes trust in collaboration?
      Observed,                                                                                Within a specific
     recognized,                                                                               scenario, setting
       history                                                                                 or understanding


             Trust is a proven contextualised product
                  of dynamic social relationships
      that can be leveraged by formal and informal rules and
                conventions within a Social Cloud
                  to facilitate as well as influence
              the scope of collaborative interchange.
                                                                                                    Implicit social
 Protocols,                                                                                          conventions
 polices etc.


21                 Participant Selection for an Experimental Social Cloud   Karlsruhe Service Research Institute
                                                                                               www.ksri.kit.edu
Use Cases
     A Social CDN                                                 A Social Volunteer Cloud
     Chard, Caton, Rana and Katz; Under                       Seminar Paper: Dominik Ernst
     Review: DataCloud @ SC 2012                                           (KIT Undergrad)

     KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




22   KIT – University of the State of Baden-Württemberg and
     National Research Center of the Helmholtz Association                        www.kit.edu
A Social Content Delivery Network for
Scientific Cooperation

                                        Replica Placement:
                                         Random
                                         Node Degree: highest
                                          no. of edges
                                         Community Node
                                          Degree (highest
                                          degree within a
                                          community, i.e. no
                                          adjacent placement)
                                         Clustering Coefficient
                                          (similar to highest
                                          betweenness scores)




                                         Karlsruhe Service Research Institute
                                                            www.ksri.kit.edu
Scenario and Community Representation




 Baseline Graph: DLBP publications graph (Kyle): 3 degrees (2009-10)
       Nodes: authors, Edges: coauthorship of 1 or more papers
    Double coauthorship: at least 2 publications
    No. of Authors: < 6 authors on the paper
    Trust: captured through prior collaborative work
    Having constructed a network, we assign replicas, and then test with
     publications from 2011
24                                                    Karlsruhe Service Research Institute
                                                                         www.ksri.kit.edu
Results (at least 60 repetitions)


Double Coauthorship                                                                                                                          No. of Coauthors

                        40                                                                                 70
                                 Random                                                                             Random

                        35       Node Degree                                                                        Node Degree
                                                                                                           60
                                 Community Node Degree                                                              Community Node Degree
                        30
                                 Clustering Coefficient                                                    50       Clustering Coefficient
 Replica Hit Rate (%)




                                                                                    Replica Hit Rate (%)
                        25
                                                                                                           40
                        20
                                                                                                           30
                        15

                                                                                                           20
                        10


                         5                                                                                 10


                         0                                                                                  0
                             1      2      3      4       5   6    7   8   9   10                               1      2      3      4       5   6      7      8      9   10
                                                  Number of Replicas                                                                 Number of Replicas



25                                                                                                                             Karlsruhe Service Research Institute
                                                                                                                                                  www.ksri.kit.edu
Social Volunteer Cloud




26                       Karlsruhe Service Research Institute
                                            www.ksri.kit.edu
Simulation-based study

 BOINC problem size: 2000
 1500 Strangers (classic VC) vs. Social Cloud of 1500:
      10 Friends, 20 Close Friends, 50 Friends, 50 Colleague, 120
       Acquaintances, 250 Community Peers, 1000 FOFs
 Strangers follow SETI@home distributions
 Social Cloud SETI@home distributions + social constructs
  to improve reliability and availability proportional to
  closeness

 Scheduler is a simple FCFS + initial performance test



27                                               Karlsruhe Service Research Institute
                                                                    www.ksri.kit.edu
Results




28        Karlsruhe Service Research Institute
                             www.ksri.kit.edu
Summary
                                                              http://www.facebook.com/SocialCloudComputing
                                                                            http://www.ksri.kit.edu/SocialCloud


     KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)




29   KIT – University of the State of Baden-Württemberg and
     National Research Center of the Helmholtz Association                                            www.kit.edu
Summary

 A Social Cloud:
    is an alternative to existing forms of distributed and collaborative
     computing
    leverages existing social relationships to act as a means to
     establish a virtual compute cloud of excess/idle resources
 We’ve looked at:
      Performance requirements of a Social Cloud
      Methods to source the Platform via the Social Cloud
      The Social and Cognitive Processes that underpin a Social Cloud
      Some Use Cases: a Social CDN and Social BOINC




                                                   Karlsruhe Service Research Institute
                                                                      www.ksri.kit.edu
Research Areas and Challenges

                       Policies
     Crowd Sourcing                      Socio-economics




Social Markets                              Strategy Proof
 & Protocols                                 Mechanisms




                                                       Remedy
     Semantics
                       Privacy                      Engineering
31                                Karlsruhe Service Research Institute
                                                     www.ksri.kit.edu
Thanks
     Come see us at eScience on Friday Oct 12th in the Application Systems and Frameworks Session:
     Haas, Caton, Trumpp, and Weinhardt; A Simulator for Social Exchanges and Collaborations -
     Architecture and Case Study


     KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI)


                                                               Social Collaboration Simulator                             Monitoring

                    User                                                  Incentive         Experiment
                                               SmallWorldNetwork                                                   ...   Application Sensor
                 Applications                                              Scheme            Controller



                  Exchange                                                                                               Exchange Sensor
                                                         Mechanism        Currency           Artifact            ...
                 Mechanisms
                                        Runtime
                                                                                                                            User Sensor
                     Core                                     User    Resource        Relationship        TrustContext
                   Elements                                                                                               Resource Sensor


32   KIT – University of the State of Baden-Württemberg and
     National Research Center of the Helmholtz Association                                                                                    www.kit.edu

Contenu connexe

Tendances

Cloud Computing? What is it and its future trends?
Cloud Computing? What is it and its future trends?Cloud Computing? What is it and its future trends?
Cloud Computing? What is it and its future trends?ziaurrehman4484
 
Cloud Computing & Cloud Brokers
Cloud Computing & Cloud Brokers Cloud Computing & Cloud Brokers
Cloud Computing & Cloud Brokers Vasan Ramadoss
 
Understanding cloud services brokerage
Understanding cloud services brokerageUnderstanding cloud services brokerage
Understanding cloud services brokerageAbel Gomez
 
Cloud computing and library services
Cloud computing and library servicesCloud computing and library services
Cloud computing and library servicesErik Mitchell
 
Cloud service brokerage explained
Cloud service brokerage explainedCloud service brokerage explained
Cloud service brokerage explainedOleksandr Varlamov
 
Cloud computing 1
Cloud computing 1Cloud computing 1
Cloud computing 1Sagar Kumar
 
Market oriented Cloud Computing
Market oriented Cloud ComputingMarket oriented Cloud Computing
Market oriented Cloud ComputingJithin Parakka
 
Master thesis presentation on 'Cloud Service Broker'
Master thesis presentation on 'Cloud Service Broker' Master thesis presentation on 'Cloud Service Broker'
Master thesis presentation on 'Cloud Service Broker' Carlos Gonçalves
 
Cloud computing for libraries an introduction
Cloud computing for libraries an introductionCloud computing for libraries an introduction
Cloud computing for libraries an introductionKrista Godfrey
 
Live Mesh Presentation Bruno Svc
Live Mesh Presentation Bruno SvcLive Mesh Presentation Bruno Svc
Live Mesh Presentation Bruno SvcWes Yanaga
 
Cloud Services Brokerage Demystified
Cloud Services Brokerage DemystifiedCloud Services Brokerage Demystified
Cloud Services Brokerage DemystifiedZach Gardner
 
Cloud Computing - FUNDAMENTALS
Cloud Computing - FUNDAMENTALSCloud Computing - FUNDAMENTALS
Cloud Computing - FUNDAMENTALSKANNANKR12
 

Tendances (20)

Forecast 2014: Cloud Service Brokering
Forecast 2014: Cloud Service BrokeringForecast 2014: Cloud Service Brokering
Forecast 2014: Cloud Service Brokering
 
Cloud service brokerage
Cloud service brokerageCloud service brokerage
Cloud service brokerage
 
Cloud Computing? What is it and its future trends?
Cloud Computing? What is it and its future trends?Cloud Computing? What is it and its future trends?
Cloud Computing? What is it and its future trends?
 
Cloud Computing & Cloud Brokers
Cloud Computing & Cloud Brokers Cloud Computing & Cloud Brokers
Cloud Computing & Cloud Brokers
 
Understanding cloud services brokerage
Understanding cloud services brokerageUnderstanding cloud services brokerage
Understanding cloud services brokerage
 
The Rise of Cloud Service Brokerage featuring Gartner and BCBS
The Rise of Cloud Service Brokerage featuring Gartner and BCBSThe Rise of Cloud Service Brokerage featuring Gartner and BCBS
The Rise of Cloud Service Brokerage featuring Gartner and BCBS
 
Cloud computing and library services
Cloud computing and library servicesCloud computing and library services
Cloud computing and library services
 
Cloud service brokerage explained
Cloud service brokerage explainedCloud service brokerage explained
Cloud service brokerage explained
 
WHAT IS CLOUD COMPUTING
WHAT IS CLOUD COMPUTINGWHAT IS CLOUD COMPUTING
WHAT IS CLOUD COMPUTING
 
Cloud computing 1
Cloud computing 1Cloud computing 1
Cloud computing 1
 
Market oriented Cloud Computing
Market oriented Cloud ComputingMarket oriented Cloud Computing
Market oriented Cloud Computing
 
Sami-Cloud
Sami-CloudSami-Cloud
Sami-Cloud
 
Master thesis presentation on 'Cloud Service Broker'
Master thesis presentation on 'Cloud Service Broker' Master thesis presentation on 'Cloud Service Broker'
Master thesis presentation on 'Cloud Service Broker'
 
Cloud computing for libraries an introduction
Cloud computing for libraries an introductionCloud computing for libraries an introduction
Cloud computing for libraries an introduction
 
Social cloud
Social cloudSocial cloud
Social cloud
 
Cloud Computing-A detailed Study
Cloud Computing-A detailed StudyCloud Computing-A detailed Study
Cloud Computing-A detailed Study
 
Cloud Computing
Cloud ComputingCloud Computing
Cloud Computing
 
Live Mesh Presentation Bruno Svc
Live Mesh Presentation Bruno SvcLive Mesh Presentation Bruno Svc
Live Mesh Presentation Bruno Svc
 
Cloud Services Brokerage Demystified
Cloud Services Brokerage DemystifiedCloud Services Brokerage Demystified
Cloud Services Brokerage Demystified
 
Cloud Computing - FUNDAMENTALS
Cloud Computing - FUNDAMENTALSCloud Computing - FUNDAMENTALS
Cloud Computing - FUNDAMENTALS
 

Similaire à Social Cloud Computing

Social Cloud talk at KSRI Service Summit 2012
Social Cloud talk at KSRI Service Summit 2012Social Cloud talk at KSRI Service Summit 2012
Social Cloud talk at KSRI Service Summit 2012Simon Caton
 
Why Libraries Virtualize
Why Libraries VirtualizeWhy Libraries Virtualize
Why Libraries VirtualizeErik Mitchell
 
Collaborative eResearch in a Social Cloud
Collaborative eResearch in a Social CloudCollaborative eResearch in a Social Cloud
Collaborative eResearch in a Social CloudSimon Caton
 
Cloud Computing in Academic Libraries A Review
Cloud Computing in Academic Libraries A ReviewCloud Computing in Academic Libraries A Review
Cloud Computing in Academic Libraries A Reviewijtsrd
 
Cloud computingjun28
Cloud computingjun28Cloud computingjun28
Cloud computingjun28korusamol
 
Cloud computing in academic libraries
Cloud computing in academic librariesCloud computing in academic libraries
Cloud computing in academic librariesErik Mitchell
 
Library discovery: past, present and some futures
Library discovery: past, present and some futuresLibrary discovery: past, present and some futures
Library discovery: past, present and some futureslisld
 
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014openi_ict
 
Cloud computing a boon for library services
Cloud computing a boon for library servicesCloud computing a boon for library services
Cloud computing a boon for library servicesKishor Satpathy
 
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaS
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaSOpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaS
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaSDaniel Krook
 
Open Source Clouds: Be The Change...
Open Source Clouds: Be The Change...Open Source Clouds: Be The Change...
Open Source Clouds: Be The Change...GreenQloud
 
Cloud computing a boon for library services
Cloud computing a boon for library servicesCloud computing a boon for library services
Cloud computing a boon for library servicesKishor Satpathy
 
Transforming Research in Collaboration with Funding Agencies
Transforming Research in Collaboration with Funding AgenciesTransforming Research in Collaboration with Funding Agencies
Transforming Research in Collaboration with Funding AgenciesAmazon Web Services
 
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"hypertext2007
 
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014Fenareti Lampathaki
 
TUW-ASE Summer 2015: IoT Cloud Systems
TUW-ASE Summer 2015:  IoT Cloud SystemsTUW-ASE Summer 2015:  IoT Cloud Systems
TUW-ASE Summer 2015: IoT Cloud SystemsHong-Linh Truong
 
Introduction to cloud Cambridge University.ppt
Introduction to cloud Cambridge University.pptIntroduction to cloud Cambridge University.ppt
Introduction to cloud Cambridge University.pptestabraqhm
 

Similaire à Social Cloud Computing (20)

Social Cloud talk at KSRI Service Summit 2012
Social Cloud talk at KSRI Service Summit 2012Social Cloud talk at KSRI Service Summit 2012
Social Cloud talk at KSRI Service Summit 2012
 
Why Libraries Virtualize
Why Libraries VirtualizeWhy Libraries Virtualize
Why Libraries Virtualize
 
Collaborative eResearch in a Social Cloud
Collaborative eResearch in a Social CloudCollaborative eResearch in a Social Cloud
Collaborative eResearch in a Social Cloud
 
Bibliotheken en cloud computing
Bibliotheken en cloud computingBibliotheken en cloud computing
Bibliotheken en cloud computing
 
Cloud Computing in Academic Libraries A Review
Cloud Computing in Academic Libraries A ReviewCloud Computing in Academic Libraries A Review
Cloud Computing in Academic Libraries A Review
 
Cloud computingjun28
Cloud computingjun28Cloud computingjun28
Cloud computingjun28
 
Cloud computingjun28
Cloud computingjun28Cloud computingjun28
Cloud computingjun28
 
Cloud computing in academic libraries
Cloud computing in academic librariesCloud computing in academic libraries
Cloud computing in academic libraries
 
Library discovery: past, present and some futures
Library discovery: past, present and some futuresLibrary discovery: past, present and some futures
Library discovery: past, present and some futures
 
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
 
Cloud computing a boon for library services
Cloud computing a boon for library servicesCloud computing a boon for library services
Cloud computing a boon for library services
 
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaS
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaSOpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaS
OpenStack and Cloud Foundry - Pair the leading open source IaaS and PaaS
 
Open Source Clouds: Be The Change...
Open Source Clouds: Be The Change...Open Source Clouds: Be The Change...
Open Source Clouds: Be The Change...
 
Cloud computing a boon for library services
Cloud computing a boon for library servicesCloud computing a boon for library services
Cloud computing a boon for library services
 
Transforming Research in Collaboration with Funding Agencies
Transforming Research in Collaboration with Funding AgenciesTransforming Research in Collaboration with Funding Agencies
Transforming Research in Collaboration with Funding Agencies
 
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"
Hypertext2007 Carole Goble Keynote - "The Return of the Prodigal Web"
 
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
OPENi perspective on APIs and Cloudlets @Athens hackathon, September 2014
 
TUW-ASE Summer 2015: IoT Cloud Systems
TUW-ASE Summer 2015:  IoT Cloud SystemsTUW-ASE Summer 2015:  IoT Cloud Systems
TUW-ASE Summer 2015: IoT Cloud Systems
 
Introduction to cloud Cambridge University.ppt
Introduction to cloud Cambridge University.pptIntroduction to cloud Cambridge University.ppt
Introduction to cloud Cambridge University.ppt
 
Introduction.ppt
Introduction.pptIntroduction.ppt
Introduction.ppt
 

Plus de Simon Caton

Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...
Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...
Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...Simon Caton
 
Research Discovery, Social Networks and VIVO
Research Discovery, Social Networks and VIVO Research Discovery, Social Networks and VIVO
Research Discovery, Social Networks and VIVO Simon Caton
 
A Social Content Delivery Network for Scientific Cooperation: Vision, Design...
A Social Content Delivery Network for Scientific Cooperation: Vision,  Design...A Social Content Delivery Network for Scientific Cooperation: Vision,  Design...
A Social Content Delivery Network for Scientific Cooperation: Vision, Design...Simon Caton
 
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...A Simulator for Social Exchanges and Collaborations - Architecture and Case S...
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...Simon Caton
 
The Gamification of Well-Being Measures
The Gamification of Well-Being MeasuresThe Gamification of Well-Being Measures
The Gamification of Well-Being MeasuresSimon Caton
 
eSoN Overview Slides
eSoN Overview SlideseSoN Overview Slides
eSoN Overview SlidesSimon Caton
 
A Social Cloud for Public eResearch
A Social Cloud for Public eResearchA Social Cloud for Public eResearch
A Social Cloud for Public eResearchSimon Caton
 
Incentivising Resource Sharing in Social Clouds
Incentivising Resource Sharing in Social CloudsIncentivising Resource Sharing in Social Clouds
Incentivising Resource Sharing in Social CloudsSimon Caton
 
Engineering Incentives in Social Clouds
Engineering Incentives in Social Clouds Engineering Incentives in Social Clouds
Engineering Incentives in Social Clouds Simon Caton
 

Plus de Simon Caton (9)

Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...
Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...
Preference-Based Resource Allocation: Using Heuristics to Solve Two-Sided Mat...
 
Research Discovery, Social Networks and VIVO
Research Discovery, Social Networks and VIVO Research Discovery, Social Networks and VIVO
Research Discovery, Social Networks and VIVO
 
A Social Content Delivery Network for Scientific Cooperation: Vision, Design...
A Social Content Delivery Network for Scientific Cooperation: Vision,  Design...A Social Content Delivery Network for Scientific Cooperation: Vision,  Design...
A Social Content Delivery Network for Scientific Cooperation: Vision, Design...
 
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...A Simulator for Social Exchanges and Collaborations - Architecture and Case S...
A Simulator for Social Exchanges and Collaborations - Architecture and Case S...
 
The Gamification of Well-Being Measures
The Gamification of Well-Being MeasuresThe Gamification of Well-Being Measures
The Gamification of Well-Being Measures
 
eSoN Overview Slides
eSoN Overview SlideseSoN Overview Slides
eSoN Overview Slides
 
A Social Cloud for Public eResearch
A Social Cloud for Public eResearchA Social Cloud for Public eResearch
A Social Cloud for Public eResearch
 
Incentivising Resource Sharing in Social Clouds
Incentivising Resource Sharing in Social CloudsIncentivising Resource Sharing in Social Clouds
Incentivising Resource Sharing in Social Clouds
 
Engineering Incentives in Social Clouds
Engineering Incentives in Social Clouds Engineering Incentives in Social Clouds
Engineering Incentives in Social Clouds
 

Dernier

🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
 
Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024The Digital Insurer
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businesspanagenda
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...Principled Technologies
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
 

Dernier (20)

🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...
Deploy with confidence: VMware Cloud Foundation 5.1 on next gen Dell PowerEdg...
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 

Social Cloud Computing

  • 1. Social Cloud Computing Simon Caton http://www.facebook.com/SocialCloudComputing http://www.ksri.kit.edu/SocialCloud KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 2. 7 years of Cloud with and the same old hurdles  Security  Lack of Customisability  Economics  Small scale consumers have ad hoc requirements  Providers have explicit incentives to lock in consumers  Countless attempts are yet to produce the open cloud market  Trust  always assumed at some level  Anonymity (Market-based/broker allocation)  Many models fall apart when this is removed Karlsruhe Service Research Institute www.ksri.kit.edu
  • 3. Collaborative (Computing) Environments  Users are lost in layer upon layer of abstraction  Often make abhorrent trust assumptions  Certificates!: Represent little more than an underlying social relationship in a dehumanized format  Massively specified – but limited in capability  Cannot see beyond the defined horizon  Usually single purpose (at most few purposes)  Examples: Karlsruhe Service Research Institute www.ksri.kit.edu
  • 4. Social Networks  Ubiquitous: Facebook surpassed 1 billion users  Represent mostly pre-existing real world relationships  Have notions of pre-existent trust fabric inherently interwoven into the network structure  Many applications now use social networks as a platform for:  Authentication e.g. Facebook Connect  Online Presence e.g. fb.com/your_page, Google Places  Application Portals e.g. progress thru processors, ASPEN and PolarGrid project Karlsruhe Service Research Institute www.ksri.kit.edu
  • 5. A Social Cloud Resources are idle 40-95% 1,000,000,000 Users On average 190 friends Users contribute to “good” causes Social Cloud: a resource, service and capability sharing framework utilizing relationships established between members of a social network Karlsruhe Service Research Institute www.ksri.kit.edu
  • 6. Talk Overview Vision of a Social Cloud Feasibility Study of a Social Market Platform Design Interaction: as a series of social and cognitive processes Use Cases: a Social CDN for academics, a social volunteer cloud Summary and Current Work 6 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 7. Social Cloud Vision Social Exchange Platform Social Cloud Karlsruhe Service Research Institute www.ksri.kit.edu
  • 8. Architecture: High Level Shared Access Ownership Social Tie “Resource” Infrastructure Server Social Middleware Karlsruhe Service Research Institute www.ksri.kit.edu
  • 9. Is a Social Cloud Feasible? From: Chard, Caton, Rana and Bubendorfer; Social Cloud: Cloud Computing in Social Networks; IEEE Cloud 2010 KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) 9 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 10. Architecture: Proof-of-concept Implementation Agreement 10 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 11. Feasibility Study: IEEE Cloud 2010  Can a Social Cloud Scale?  What are the computational requirements for an avg. SN?  Can a Social Cloud function in a timely manner as a Facebook application? Run on a single desktop 11 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 12. Constructing a Social Cloud Platform From: Haas, Caton, Chard and Weinhardt; Co-Operative Infrastructures: An Economic Model for Providing Infrastructures for Social Cloud Computing; HICSS 2013 KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) 12 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 13. How do we build a platform for a Social Cloud? Platform needs resources to:  Calculate allocations  Save bids and asks  Answer queries  … 13 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 14. Platform Design via Co-op Infrastructures A co-op is a scalable computing platform where all (computational) resources constituting the platform's infrastructure, as well as those made available over the platform, are owned and/or managed by its users. What contribution schemes can secure the resources are needed to keep a platform accessible? 14 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 15. Setting  Platform Load: interpolated version of Cloud paper for a range of users [10, 400]  Resource availability via SETI@home user distributions  Users are modeled with varied compute resources  Put these together and we “know” the total contribution needed  Load is almost worst case:  every trade occurs nearly in parallel  half the social cloud takes part (if they meet min requirements)  we include simple levels of redundancy 15 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 16. Contribution Percentage 0.9 Availability Percentage 0.4 0.8 0.35 0.7 0.3 0.6 Fixed Contribution 0.5 0.4 0.25 0.2 0.3 0.15 0.2 0.1  Idea: Users have to provide a given percentage of their available resources to 0.1 0.05 0 0 the co-operative infrastructure 400 10 20 50 100 200 10 20 50 100 200 400  Questions: Number of Users Number of Users rho_star do we set the percentage? How rho_star*1.1 rho_star*1.2 rho_star*1.5 rho_star rho_star*1.1 rho_star*1.2 rho_star*1.5 a) System the effect of the contribution percentage on system reliability? of users, worst What is availability, worst case b) Average contribution requirements case requirements 1 0.45 Contribution Percentage 0.9 Availability Percentage 0.4 0.8 0.35 0.7 0.3 0.6 0.5 0.25 0.4 0.2 0.3 0.15 0.2 0.1 0.1 0.05 0 0 10 20 50 100 200 400 10 20 50 100 200 400 Number of Users Number of Users rho_star rho_star*1.1 rho_star*1.2 rho_star*1.5 rho_star rho_star*1.1 rho_star*1.2 rho_star*1.5 d) System availability, average e) Average contribution of users, Platform Availability Av. % Contribution requirements average requirements 16 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 17. Voluntary Contribution  Idea: Let users choose the amount of resources they contribute, based on individual preferences  Approach:  User behavior modeled through Utility Functions with Other-Regarding Preferences (User Types: self-interested, altruist, hybrid)  Study dependence of system performance on the distribution of user types  Price variable to capture relative ease/difficulty to provide resources Platform Availability Av. % Contribution 17 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 18. Interaction in a Social Cloud as Social and Cognitive Processes From: Caton, Dukat, Grenz, Haas, Pfadenhauer and Weinhardt; Foundations of Trust: Contextualising Trust in Social Clouds; IEEE Social Computing and Its Applications 2012 KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) 18 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 19. Interaction: Social and Cognitive Processes Social Cloud Ex-Ante Social Ex-Post • Motivation: Outcome, Social Context, History Interchange • Feedback: Locally and to Network(s) • Excess driven demand • Formal Processes: • Recommendations: Rewards • Demand induced Social Initialisation, Identification, and Sanctions Capital Allocation, Provisioning • Interaction Archiving • Informal Social (History) Communication and Coordination Prior Completion Expectations Evolution of Relationship(s) 19 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 20.  … an attribute of a relationship  … in the necessary competence to be able to deliver  … to deliver, keep promises etc. 20 Participant Selection for an Experimental Social Cloud Karlsruhe Service Research Institute www.ksri.kit.edu
  • 21. What characterizes trust in collaboration? Observed, Within a specific recognized, scenario, setting history or understanding Trust is a proven contextualised product of dynamic social relationships that can be leveraged by formal and informal rules and conventions within a Social Cloud to facilitate as well as influence the scope of collaborative interchange. Implicit social Protocols, conventions polices etc. 21 Participant Selection for an Experimental Social Cloud Karlsruhe Service Research Institute www.ksri.kit.edu
  • 22. Use Cases A Social CDN A Social Volunteer Cloud Chard, Caton, Rana and Katz; Under Seminar Paper: Dominik Ernst Review: DataCloud @ SC 2012 (KIT Undergrad) KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) 22 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 23. A Social Content Delivery Network for Scientific Cooperation Replica Placement:  Random  Node Degree: highest no. of edges  Community Node Degree (highest degree within a community, i.e. no adjacent placement)  Clustering Coefficient (similar to highest betweenness scores) Karlsruhe Service Research Institute www.ksri.kit.edu
  • 24. Scenario and Community Representation  Baseline Graph: DLBP publications graph (Kyle): 3 degrees (2009-10)  Nodes: authors, Edges: coauthorship of 1 or more papers  Double coauthorship: at least 2 publications  No. of Authors: < 6 authors on the paper  Trust: captured through prior collaborative work  Having constructed a network, we assign replicas, and then test with publications from 2011 24 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 25. Results (at least 60 repetitions) Double Coauthorship No. of Coauthors 40 70 Random Random 35 Node Degree Node Degree 60 Community Node Degree Community Node Degree 30 Clustering Coefficient 50 Clustering Coefficient Replica Hit Rate (%) Replica Hit Rate (%) 25 40 20 30 15 20 10 5 10 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Number of Replicas Number of Replicas 25 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 26. Social Volunteer Cloud 26 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 27. Simulation-based study  BOINC problem size: 2000  1500 Strangers (classic VC) vs. Social Cloud of 1500:  10 Friends, 20 Close Friends, 50 Friends, 50 Colleague, 120 Acquaintances, 250 Community Peers, 1000 FOFs  Strangers follow SETI@home distributions  Social Cloud SETI@home distributions + social constructs to improve reliability and availability proportional to closeness  Scheduler is a simple FCFS + initial performance test 27 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 28. Results 28 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 29. Summary http://www.facebook.com/SocialCloudComputing http://www.ksri.kit.edu/SocialCloud KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) 29 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu
  • 30. Summary  A Social Cloud:  is an alternative to existing forms of distributed and collaborative computing  leverages existing social relationships to act as a means to establish a virtual compute cloud of excess/idle resources  We’ve looked at:  Performance requirements of a Social Cloud  Methods to source the Platform via the Social Cloud  The Social and Cognitive Processes that underpin a Social Cloud  Some Use Cases: a Social CDN and Social BOINC Karlsruhe Service Research Institute www.ksri.kit.edu
  • 31. Research Areas and Challenges Policies Crowd Sourcing Socio-economics Social Markets Strategy Proof & Protocols Mechanisms Remedy Semantics Privacy Engineering 31 Karlsruhe Service Research Institute www.ksri.kit.edu
  • 32. Thanks Come see us at eScience on Friday Oct 12th in the Application Systems and Frameworks Session: Haas, Caton, Trumpp, and Weinhardt; A Simulator for Social Exchanges and Collaborations - Architecture and Case Study KARLSRUHE SERVICE RESEARCH INSTITUTE (KSRI) Social Collaboration Simulator Monitoring User Incentive Experiment SmallWorldNetwork ... Application Sensor Applications Scheme Controller Exchange Exchange Sensor Mechanism Currency Artifact ... Mechanisms Runtime User Sensor Core User Resource Relationship TrustContext Elements Resource Sensor 32 KIT – University of the State of Baden-Württemberg and National Research Center of the Helmholtz Association www.kit.edu

Notes de l'éditeur

  1. Many service types: people servicesComputational servicesStorage servicesIn industry solutions this assumption is typically replaced by consumer proactive/reactive action – e.g. Amazon SLAOr they do not implement enforcement policiesThe problem:Anonymity between participants is commonE.g. Allocation through auctions or other market mechanismsThe models fall apart completely if this assumption is removed
  2. Social Networks model relationshipsOpenSocial &amp; OpenId, used by most social networking sites, andFacebook’s bespoke application framework
  3. Economic model:Allows representation of User preferences as utility functionsCaptures things like costs to contribute as relative variables
  4. Graph Nodes Publications EdgesBaseline 2335 1163 17973Double-Author 811 1163 5123Number of Authors 604 435 1988