SlideShare a Scribd company logo
1 of 27
Utility Driven Service Routing
over Large Scale Infrastructures


Pablo Chacin
Polytechnic University of Catalonia
(UPC), Spain
Authors
• Pablo Chacin, Polytechnic University of
Catalonia, Spain (UPC)
• Leandro Navarro, UPC
• Pedro Garcia López, Rovira i Virgili
University, Spain
Key Points
   • UDON is an Utility Driven Overlay Network for routing
     service requests to service instances that match some QoS
     requirements
   • It is aimed for highly dynamic large-scale shared
     infrastructures.
   • Combines an application provided utility function to express
     QoS with an epidemic protocol to disseminate the
     information that supports the routing
   • Experimental analysis shows that UDON allocates requests
     meeting QoS with a high probability and low overhead; it is
     scalable, robust and adapts well to a wide range of
     conditions.


13-15 December 2010     ServiceWave 2010
Outline
   •   Defining the problem context
   •   Design principles
   •   Experimental evaluation
   •   Conclusions




13-15 December 2010   ServiceWave 2010
Internet of Services




        Source: Schroth, C., Janner, T.: Web 2.0 and soa: Converging concepts enabling
        the internet of services. IT Professional 9(3), 36–41 (May/June 2007)
13-15 December 2010           ServiceWave 2010
Service Deployment




13-15 December 2010   ServiceWave 2010
Challenges
   • Non dedicated Servers
      – The QoS a server can offer is hard to predict
   • Fluctuations in the demand
   • Different QoS requirements for different users
      – e.g. free/paid; bronze/silver/gold
   • Large scale
   • Number of instances may vary
      – Activations/deactivations due to fluctuations on the
        demand
      – Failures


13-15 December 2010     ServiceWave 2010
Guiding principles
   • Decentralized decisions using local information
      – No global view; no single point of failure; more
        scalable and adaptable
   • Representation of QoS as an Utility Function
      – Compact representation
      – Facilitate comparisons despite heterogeneity
   • Model-less adaptation
     – No need to elicit or learn a performance model for
       the systems
     – If information is not exact, rationality may not
       help.

13-15 December 2010   ServiceWave 2010
System Model




13-15 December 2010    ServiceWave 2010
Utility Function
   • In economics, utility is a
     measure of relative
     satisfaction
   • Summarizes multiple
     attributes into a single
     scalar value
      – F(a1,..an) → [0,1]
   • Facilitates comparison,
     allow private evaluations
                                           Cobb-Douglas utility function
                                           U(t,c) = t(ac(1-a)
                                           t = execution time
                                           c = cost
13-15 December 2010     ServiceWave 2010
Epidemic Overlay
   • Simple maintenance algorithm
      – Each node has a local view of
        the state of a set of neighbors
      – Periodically choses some
        neighbors and sends its local
        view + own state
      – Each node merges its local
        view with the received views
        keeping the most recently
        updated entries
   • Disseminates information with low
     overhead
   • Highly scalable and resilient


13-15 December 2010     ServiceWave 2010
Randomized Greedy Utility
              Routing
   • Multi-hop routing using local
     information
      – On each hop, ranks
         neighbors based on its
         (potentially outdated)
         utility
      – Forward to the node with
         a probability based on
         ranking
   • Simple concept. Allows
     multiple heuristics for          Image source: physics.org
     ranking (evaluation is an        Greedy Routing Enables Network Navigation
                                      Without a 'Map'
     ongoing work)                    http://www.physorg.com/news154093231.html



13-15 December 2010     ServiceWave 2010
Evaluation




13-15 December 2010   ServiceWave 2010
Simulation Model
   • Network topology is abstracted
      – One single cluster, 1000's of servers.
      – Constant, negligible delays
   • Utility Function simulated as a Random Process
      – Make evaluation more general, not tied to a
        particular utility definition
      – Evaluate the effect of different parameters
   • Compared with other overlays of the same family
      – Random: no organization (baseline)
      – Gradient: keep instances with similar QoS close

13-15 December 2010   ServiceWave 2010
The Simulation of the Utility
             Function




13-15 December 2010   ServiceWave 2010
Metrics
   • Overlay (information dissemination)
      – Age: how old is the information in the
        local view (average)
      – Staleness: how accurate is the local view
        with respect of real current information
   • Routing
      – Satisfied demand: how effective and
        reliable is the allocation (% of success)
      – Hops: how efficient
13-15 December 2010   ServiceWave 2010
Overlay
                                         Maintains “fresh”
                                         information




                                         Minimizes
                                         staleness



13-15 December 2010   ServiceWave 2010
Performance




 Tolerance: maximum allowed difference
 between required QoS and node's utility:       Allocates requests with high
 ~ 1.0 any node with a higher utility matches   probability, and low number or
 ~ 0.0 only node with the exact demanded
 utility matches                                hops, even under very
                                                demanding search criteria (low
                                                tolerance)
13-15 December 2010              ServiceWave 2010
Performance looking for
            scarce resources
                                         Allocates requests
                                         even when target
                                         nodes are scarce.




13-15 December 2010   ServiceWave 2010
Churn

                                         Performance
                                         “gracefully” degrades
                                         under high churn




13-15 December 2010   ServiceWave 2010
Variation in Utility




                      Allocates requests even under
                      highly fluctuating conditions.


13-15 December 2010       ServiceWave 2010
Sensitivity to Operational
              Parameters




           Optimal setup demands low
           communication overhead

13-15 December 2010   ServiceWave 2010
Discussion




13-15 December 2010   ServiceWave 2010
Conclusions
   • Simple, principled solution for routing requests
     over large-scale cluster-based web services on
     shared infrastructures
   • UDON meets requirements on scenarios of
     interest and shows desirable properties
      – Effective
      – Low overhead
      – Scalable
      – Very adaptable
      – Robust

13-15 December 2010    ServiceWave 2010
(Near) Future work
   • Apply UDON to A concrete scenario
      – Simulated cluster based web services
      – Use concrete utility functions
   • Evaluate alternative routing heuristics
   • Propagate information based on usefulness:
     see which QoS are more demanded and
     propagate information of nodes that offer it
     with higher probability
   • Consider locality when selecting neighbors to
     adapt to wide area distributed clusters (multi-
     site)
13-15 December 2010   ServiceWave 2010
Questions? . . . Thanks.

                      pchacin@ac.upc.edu



13-15 December 2010      ServiceWave 2010
ICSOC-ServiceWave 2009

More Related Content

Similar to Pablo Chacin (Polytechnic University of Catalonia, Spain): Utility Driven Service Routing over Large Scale Infrastructures

Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Pradeeban Kathiravelu, Ph.D.
 
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreeThe UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreePradeeban Kathiravelu, Ph.D.
 
Introduction(2)
Introduction(2)Introduction(2)
Introduction(2)trayyoo
 
PacketCloud: an Open Platform for Elastic In-network Services.
PacketCloud: an Open Platform for Elastic In-network Services. PacketCloud: an Open Platform for Elastic In-network Services.
PacketCloud: an Open Platform for Elastic In-network Services. yeung2000
 
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...Tal Lavian Ph.D.
 
20-datacenter-measurements.pptx
20-datacenter-measurements.pptx20-datacenter-measurements.pptx
20-datacenter-measurements.pptxSteve491226
 
Introduction to SDN: Software Defined Networking
Introduction to SDN: Software Defined NetworkingIntroduction to SDN: Software Defined Networking
Introduction to SDN: Software Defined NetworkingAnkita Mahajan
 
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...Nuage Networks
 
Next Generation Optical Networking: Software-Defined Optical Networking
Next Generation Optical Networking: Software-Defined Optical NetworkingNext Generation Optical Networking: Software-Defined Optical Networking
Next Generation Optical Networking: Software-Defined Optical NetworkingADVA
 
Self-Tuning and Managing Services
Self-Tuning and Managing ServicesSelf-Tuning and Managing Services
Self-Tuning and Managing ServicesReza Rahimi
 
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...Tal Lavian Ph.D.
 
Nuage meetup - Flexible and agile Software Defined Networking (SDN)
Nuage meetup - Flexible and agile Software Defined Networking (SDN)Nuage meetup - Flexible and agile Software Defined Networking (SDN)
Nuage meetup - Flexible and agile Software Defined Networking (SDN)SDN_Paris
 
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...Tal Lavian Ph.D.
 
Services and applications’ infrastructure for agile optical networks
Services and applications’ infrastructure for agile optical networksServices and applications’ infrastructure for agile optical networks
Services and applications’ infrastructure for agile optical networksTal Lavian Ph.D.
 
NetScout nGeniusONE overview
NetScout nGeniusONE overviewNetScout nGeniusONE overview
NetScout nGeniusONE overviewBAKOTECH
 

Similar to Pablo Chacin (Polytechnic University of Catalonia, Spain): Utility Driven Service Routing over Large Scale Infrastructures (20)

Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...
 
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreeThe UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
 
RTI Technical Road Show SPAWAR SD
RTI Technical Road Show SPAWAR SDRTI Technical Road Show SPAWAR SD
RTI Technical Road Show SPAWAR SD
 
Qo s rsvp......
Qo s rsvp......Qo s rsvp......
Qo s rsvp......
 
Introduction(2)
Introduction(2)Introduction(2)
Introduction(2)
 
PacketCloud: an Open Platform for Elastic In-network Services.
PacketCloud: an Open Platform for Elastic In-network Services. PacketCloud: an Open Platform for Elastic In-network Services.
PacketCloud: an Open Platform for Elastic In-network Services.
 
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
 
20-datacenter-measurements.pptx
20-datacenter-measurements.pptx20-datacenter-measurements.pptx
20-datacenter-measurements.pptx
 
Introduction to SDN: Software Defined Networking
Introduction to SDN: Software Defined NetworkingIntroduction to SDN: Software Defined Networking
Introduction to SDN: Software Defined Networking
 
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...
Alcatel-Lucent Tech Symposium 2013: SDN: Innovating for Growth with Nuage Net...
 
Next Generation Optical Networking: Software-Defined Optical Networking
Next Generation Optical Networking: Software-Defined Optical NetworkingNext Generation Optical Networking: Software-Defined Optical Networking
Next Generation Optical Networking: Software-Defined Optical Networking
 
Alternative metrics
Alternative metricsAlternative metrics
Alternative metrics
 
IJET-V3I1P13
IJET-V3I1P13IJET-V3I1P13
IJET-V3I1P13
 
Self-Tuning and Managing Services
Self-Tuning and Managing ServicesSelf-Tuning and Managing Services
Self-Tuning and Managing Services
 
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...
 
Networked 3-D Virtual Collaboration in Science and Education: Towards 'Web 3....
Networked 3-D Virtual Collaboration in Science and Education: Towards 'Web 3....Networked 3-D Virtual Collaboration in Science and Education: Towards 'Web 3....
Networked 3-D Virtual Collaboration in Science and Education: Towards 'Web 3....
 
Nuage meetup - Flexible and agile Software Defined Networking (SDN)
Nuage meetup - Flexible and agile Software Defined Networking (SDN)Nuage meetup - Flexible and agile Software Defined Networking (SDN)
Nuage meetup - Flexible and agile Software Defined Networking (SDN)
 
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...
A Platform for Large-Scale Grid Data Service on Dynamic High-Performance Netw...
 
Services and applications’ infrastructure for agile optical networks
Services and applications’ infrastructure for agile optical networksServices and applications’ infrastructure for agile optical networks
Services and applications’ infrastructure for agile optical networks
 
NetScout nGeniusONE overview
NetScout nGeniusONE overviewNetScout nGeniusONE overview
NetScout nGeniusONE overview
 

More from ServiceWave 2010

Massonet Philippe Panel - Security in the clouds: An Academic Perspective
Massonet Philippe Panel - Security in the clouds: An Academic PerspectiveMassonet Philippe Panel - Security in the clouds: An Academic Perspective
Massonet Philippe Panel - Security in the clouds: An Academic PerspectiveServiceWave 2010
 
Rainer Zimmermann (European Commission): The role of the European Commission ...
Rainer Zimmermann (European Commission): The role of the European Commission ...Rainer Zimmermann (European Commission): The role of the European Commission ...
Rainer Zimmermann (European Commission): The role of the European Commission ...ServiceWave 2010
 
Usman Wajid: Service-based Application Development by Ordinary End Users and ...
Usman Wajid: Service-based Application Development by Ordinary End Users and ...Usman Wajid: Service-based Application Development by Ordinary End Users and ...
Usman Wajid: Service-based Application Development by Ordinary End Users and ...ServiceWave 2010
 
D. Meiländer, S. Gorlatch, C. Cappiello, V. Mazza, R. Kazhamiakin, and A. Buc...
D. Meiländer, S. Gorlatch, C. Cappiello,V. Mazza, R. Kazhamiakin, and A. Buc...D. Meiländer, S. Gorlatch, C. Cappiello,V. Mazza, R. Kazhamiakin, and A. Buc...
D. Meiländer, S. Gorlatch, C. Cappiello, V. Mazza, R. Kazhamiakin, and A. Buc...ServiceWave 2010
 
Maryam Razavian: A Frame of Reference for SOA Migration
Maryam Razavian: A Frame of Reference for SOA MigrationMaryam Razavian: A Frame of Reference for SOA Migration
Maryam Razavian: A Frame of Reference for SOA MigrationServiceWave 2010
 
Security in the Clouds Panel Chair: Mike Surridge
Security in the Clouds Panel Chair: Mike SurridgeSecurity in the Clouds Panel Chair: Mike Surridge
Security in the Clouds Panel Chair: Mike SurridgeServiceWave 2010
 
Martine Lapierre - Security in Cloud computing: sharing more than resources
Martine Lapierre - Security in Cloud computing: sharing more than resourcesMartine Lapierre - Security in Cloud computing: sharing more than resources
Martine Lapierre - Security in Cloud computing: sharing more than resourcesServiceWave 2010
 
Chen Wang, Pazat, Di Napoli, Giordano: A Chemical Based Middleware for Workf...
Chen Wang, Pazat, Di Napoli, Giordano:  A Chemical Based Middleware for Workf...Chen Wang, Pazat, Di Napoli, Giordano:  A Chemical Based Middleware for Workf...
Chen Wang, Pazat, Di Napoli, Giordano: A Chemical Based Middleware for Workf...ServiceWave 2010
 
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...ServiceWave 2010
 
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...ServiceWave 2010
 
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...ServiceWave 2010
 
Scott Kirkpatrick (Hebrew University): OneLab: Federation and Testbeds
Scott Kirkpatrick (Hebrew University): OneLab: Federation and TestbedsScott Kirkpatrick (Hebrew University): OneLab: Federation and Testbeds
Scott Kirkpatrick (Hebrew University): OneLab: Federation and TestbedsServiceWave 2010
 
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...ServiceWave 2010
 
Orestis Terzidis - Taking the Internet of Services to the Next Level
Orestis Terzidis - Taking the Internet of Services to the Next LevelOrestis Terzidis - Taking the Internet of Services to the Next Level
Orestis Terzidis - Taking the Internet of Services to the Next LevelServiceWave 2010
 
Martine Lapierre - Security & Privacy trends for Urban & transport applications
Martine Lapierre - Security & Privacy trends for Urban & transport applications Martine Lapierre - Security & Privacy trends for Urban & transport applications
Martine Lapierre - Security & Privacy trends for Urban & transport applications ServiceWave 2010
 
Mário Campolargo - Services and clouds as cornerstones of the Digital Agenda
Mário Campolargo - Services and clouds as cornerstones of the Digital AgendaMário Campolargo - Services and clouds as cornerstones of the Digital Agenda
Mário Campolargo - Services and clouds as cornerstones of the Digital AgendaServiceWave 2010
 

More from ServiceWave 2010 (20)

03 v pevtschin
03 v pevtschin03 v pevtschin
03 v pevtschin
 
03 v pevtschin
03 v pevtschin03 v pevtschin
03 v pevtschin
 
Massonet Philippe Panel - Security in the clouds: An Academic Perspective
Massonet Philippe Panel - Security in the clouds: An Academic PerspectiveMassonet Philippe Panel - Security in the clouds: An Academic Perspective
Massonet Philippe Panel - Security in the clouds: An Academic Perspective
 
Rainer Zimmermann (European Commission): The role of the European Commission ...
Rainer Zimmermann (European Commission): The role of the European Commission ...Rainer Zimmermann (European Commission): The role of the European Commission ...
Rainer Zimmermann (European Commission): The role of the European Commission ...
 
Usman Wajid: Service-based Application Development by Ordinary End Users and ...
Usman Wajid: Service-based Application Development by Ordinary End Users and ...Usman Wajid: Service-based Application Development by Ordinary End Users and ...
Usman Wajid: Service-based Application Development by Ordinary End Users and ...
 
D. Meiländer, S. Gorlatch, C. Cappiello, V. Mazza, R. Kazhamiakin, and A. Buc...
D. Meiländer, S. Gorlatch, C. Cappiello,V. Mazza, R. Kazhamiakin, and A. Buc...D. Meiländer, S. Gorlatch, C. Cappiello,V. Mazza, R. Kazhamiakin, and A. Buc...
D. Meiländer, S. Gorlatch, C. Cappiello, V. Mazza, R. Kazhamiakin, and A. Buc...
 
Maryam Razavian: A Frame of Reference for SOA Migration
Maryam Razavian: A Frame of Reference for SOA MigrationMaryam Razavian: A Frame of Reference for SOA Migration
Maryam Razavian: A Frame of Reference for SOA Migration
 
Security in the Clouds Panel Chair: Mike Surridge
Security in the Clouds Panel Chair: Mike SurridgeSecurity in the Clouds Panel Chair: Mike Surridge
Security in the Clouds Panel Chair: Mike Surridge
 
Martine Lapierre - Security in Cloud computing: sharing more than resources
Martine Lapierre - Security in Cloud computing: sharing more than resourcesMartine Lapierre - Security in Cloud computing: sharing more than resources
Martine Lapierre - Security in Cloud computing: sharing more than resources
 
Chen Wang, Pazat, Di Napoli, Giordano: A Chemical Based Middleware for Workf...
Chen Wang, Pazat, Di Napoli, Giordano:  A Chemical Based Middleware for Workf...Chen Wang, Pazat, Di Napoli, Giordano:  A Chemical Based Middleware for Workf...
Chen Wang, Pazat, Di Napoli, Giordano: A Chemical Based Middleware for Workf...
 
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...
Roman Khazankin (Vienna University of Technology): Providence: A Framework fo...
 
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...
Maurer, Sakellariou, Brandic : Simulating Autonomic SLA Enactment in Clouds u...
 
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...
Andreas Wolke: TwoSpot. A Cloud Platform for Scaling out Web Applications dyn...
 
Scott Kirkpatrick (Hebrew University): OneLab: Federation and Testbeds
Scott Kirkpatrick (Hebrew University): OneLab: Federation and TestbedsScott Kirkpatrick (Hebrew University): OneLab: Federation and Testbeds
Scott Kirkpatrick (Hebrew University): OneLab: Federation and Testbeds
 
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...
Jacques Magen (FIRESTATION): Testbeds for Service Deployment. FIRESTATION’s v...
 
1 sw2010 testbeds-panel
1  sw2010 testbeds-panel1  sw2010 testbeds-panel
1 sw2010 testbeds-panel
 
Orestis Terzidis - Taking the Internet of Services to the Next Level
Orestis Terzidis - Taking the Internet of Services to the Next LevelOrestis Terzidis - Taking the Internet of Services to the Next Level
Orestis Terzidis - Taking the Internet of Services to the Next Level
 
Martine Lapierre - Security & Privacy trends for Urban & transport applications
Martine Lapierre - Security & Privacy trends for Urban & transport applications Martine Lapierre - Security & Privacy trends for Urban & transport applications
Martine Lapierre - Security & Privacy trends for Urban & transport applications
 
Mário Campolargo - Services and clouds as cornerstones of the Digital Agenda
Mário Campolargo - Services and clouds as cornerstones of the Digital AgendaMário Campolargo - Services and clouds as cornerstones of the Digital Agenda
Mário Campolargo - Services and clouds as cornerstones of the Digital Agenda
 
Fire Demos
Fire DemosFire Demos
Fire Demos
 

Pablo Chacin (Polytechnic University of Catalonia, Spain): Utility Driven Service Routing over Large Scale Infrastructures

  • 1. Utility Driven Service Routing over Large Scale Infrastructures Pablo Chacin Polytechnic University of Catalonia (UPC), Spain
  • 2. Authors • Pablo Chacin, Polytechnic University of Catalonia, Spain (UPC) • Leandro Navarro, UPC • Pedro Garcia López, Rovira i Virgili University, Spain
  • 3. Key Points • UDON is an Utility Driven Overlay Network for routing service requests to service instances that match some QoS requirements • It is aimed for highly dynamic large-scale shared infrastructures. • Combines an application provided utility function to express QoS with an epidemic protocol to disseminate the information that supports the routing • Experimental analysis shows that UDON allocates requests meeting QoS with a high probability and low overhead; it is scalable, robust and adapts well to a wide range of conditions. 13-15 December 2010 ServiceWave 2010
  • 4. Outline • Defining the problem context • Design principles • Experimental evaluation • Conclusions 13-15 December 2010 ServiceWave 2010
  • 5. Internet of Services Source: Schroth, C., Janner, T.: Web 2.0 and soa: Converging concepts enabling the internet of services. IT Professional 9(3), 36–41 (May/June 2007) 13-15 December 2010 ServiceWave 2010
  • 6. Service Deployment 13-15 December 2010 ServiceWave 2010
  • 7. Challenges • Non dedicated Servers – The QoS a server can offer is hard to predict • Fluctuations in the demand • Different QoS requirements for different users – e.g. free/paid; bronze/silver/gold • Large scale • Number of instances may vary – Activations/deactivations due to fluctuations on the demand – Failures 13-15 December 2010 ServiceWave 2010
  • 8. Guiding principles • Decentralized decisions using local information – No global view; no single point of failure; more scalable and adaptable • Representation of QoS as an Utility Function – Compact representation – Facilitate comparisons despite heterogeneity • Model-less adaptation – No need to elicit or learn a performance model for the systems – If information is not exact, rationality may not help. 13-15 December 2010 ServiceWave 2010
  • 9. System Model 13-15 December 2010 ServiceWave 2010
  • 10. Utility Function • In economics, utility is a measure of relative satisfaction • Summarizes multiple attributes into a single scalar value – F(a1,..an) → [0,1] • Facilitates comparison, allow private evaluations Cobb-Douglas utility function U(t,c) = t(ac(1-a) t = execution time c = cost 13-15 December 2010 ServiceWave 2010
  • 11. Epidemic Overlay • Simple maintenance algorithm – Each node has a local view of the state of a set of neighbors – Periodically choses some neighbors and sends its local view + own state – Each node merges its local view with the received views keeping the most recently updated entries • Disseminates information with low overhead • Highly scalable and resilient 13-15 December 2010 ServiceWave 2010
  • 12. Randomized Greedy Utility Routing • Multi-hop routing using local information – On each hop, ranks neighbors based on its (potentially outdated) utility – Forward to the node with a probability based on ranking • Simple concept. Allows multiple heuristics for Image source: physics.org ranking (evaluation is an Greedy Routing Enables Network Navigation Without a 'Map' ongoing work) http://www.physorg.com/news154093231.html 13-15 December 2010 ServiceWave 2010
  • 13. Evaluation 13-15 December 2010 ServiceWave 2010
  • 14. Simulation Model • Network topology is abstracted – One single cluster, 1000's of servers. – Constant, negligible delays • Utility Function simulated as a Random Process – Make evaluation more general, not tied to a particular utility definition – Evaluate the effect of different parameters • Compared with other overlays of the same family – Random: no organization (baseline) – Gradient: keep instances with similar QoS close 13-15 December 2010 ServiceWave 2010
  • 15. The Simulation of the Utility Function 13-15 December 2010 ServiceWave 2010
  • 16. Metrics • Overlay (information dissemination) – Age: how old is the information in the local view (average) – Staleness: how accurate is the local view with respect of real current information • Routing – Satisfied demand: how effective and reliable is the allocation (% of success) – Hops: how efficient 13-15 December 2010 ServiceWave 2010
  • 17. Overlay Maintains “fresh” information Minimizes staleness 13-15 December 2010 ServiceWave 2010
  • 18. Performance Tolerance: maximum allowed difference between required QoS and node's utility: Allocates requests with high ~ 1.0 any node with a higher utility matches probability, and low number or ~ 0.0 only node with the exact demanded utility matches hops, even under very demanding search criteria (low tolerance) 13-15 December 2010 ServiceWave 2010
  • 19. Performance looking for scarce resources Allocates requests even when target nodes are scarce. 13-15 December 2010 ServiceWave 2010
  • 20. Churn Performance “gracefully” degrades under high churn 13-15 December 2010 ServiceWave 2010
  • 21. Variation in Utility Allocates requests even under highly fluctuating conditions. 13-15 December 2010 ServiceWave 2010
  • 22. Sensitivity to Operational Parameters Optimal setup demands low communication overhead 13-15 December 2010 ServiceWave 2010
  • 23. Discussion 13-15 December 2010 ServiceWave 2010
  • 24. Conclusions • Simple, principled solution for routing requests over large-scale cluster-based web services on shared infrastructures • UDON meets requirements on scenarios of interest and shows desirable properties – Effective – Low overhead – Scalable – Very adaptable – Robust 13-15 December 2010 ServiceWave 2010
  • 25. (Near) Future work • Apply UDON to A concrete scenario – Simulated cluster based web services – Use concrete utility functions • Evaluate alternative routing heuristics • Propagate information based on usefulness: see which QoS are more demanded and propagate information of nodes that offer it with higher probability • Consider locality when selecting neighbors to adapt to wide area distributed clusters (multi- site) 13-15 December 2010 ServiceWave 2010
  • 26. Questions? . . . Thanks. pchacin@ac.upc.edu 13-15 December 2010 ServiceWave 2010