advertisement

Collaborative Task Execution In Volunteer Clouds (or how to choose a sub-reviewer)

50 %
50 %
advertisement
Information about Collaborative Task Execution In Volunteer Clouds (or how to choose a...
Technology

Published on February 20, 2014

Author: albertolluch

Source: slideshare.net

Description

My talk at the 2nd General Meeting of the CINA project, Bologna, 18-20 Feb 2014.

The increasing diffusion of cloud technologies offers new opportunities for distributed and collaborative computing. Volunteer clouds are a prominent example, where participants join and leave the platform and collaborate by sharing computational resources. The high complexity, dynamism and unpredictability of such scenarios call for decentralized self-* approaches. We present in this paper a framework for the design and evaluation of self-adaptive collaborative task execution strategies in volunteer clouds. As a byproduct, we propose a novel strategy based on the Ant Colony Optimization paradigm, that we validate through simulation-based statistical analysis over Google workload data.
advertisement

Collaborative Task Execution In Volunteer Clouds -- Michele Amoretti, PARMA -- Alberto Lluch Lafuente, IMT -- Stefano Sebastio, IMT 2nd General Meeting, Bologna, 18-20 Feb 2014

Collaborative Task Execution in Volunteer Clouds -- Michele Amoreti, PARMA -- Alberto Lluch Lafuente, IMT -- Stefano Sebastio, IMT

Collaborative Task Execution in Volunteer Clouds -- Michele Amoreti, PARMA -- Alberto Lluch Lafuente, IMT -- Stefano Sebastio, IMT

c ho o s e ow to cho ose H ow to H Collaborativeewer Execution in ub-reviiewer s ub-rev Task a s a Volunteer Clouds -- Michele Amoreti, PARMA -- Alberto Lluch Lafuente, IMT -- Stefano Sebastio, IMT

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 X X X X X X X X X

ewers e revi hoos c domly t) ran lmos (a

SciFi Community

SciFi Community 1 Unstructured network

SciFi Community 1 Unstructured network 2 All members generate review tasks

SciFi Community 1 Unstructured network 2 All members generate review tasks 3 All members perform reviews

SciFi Community 1 Unstructured network 2 All members generate review tasks 3 All members perform reviews 4 Review requests may be forwarded

SciFi Community 1 Unstructured network 2 All members generate review tasks 3 All members perform reviews 4 Review requests may be forwarded 5 All members apply the same algorithm

SciFi Reviewers

SciFi Reviewers 1 No rescheduling, no priorities.

SciFi Reviewers 1 No rescheduling, no priorities. 2 Accept request iff CoS met.

SciFi Reviewers 1 No rescheduling, no priorities. 2 Accept request iff CoS met. 3 No delays.

SciFi Reviewers 1 No rescheduling, no priorities. 2 Accept request iff CoS met. 3 No delays. 4 Reply/Forward requests immediately.

SciFi Reviewers 1 No rescheduling, no priorities. 2 Accept request iff CoS met. 3 No delays. 4 Reply/Forward requests immediately. 5 Disclose confidence on research topics.

ALGORITHM 1: RANDOM

Algorithm 1: Random Each outgoing arc has the same probability of being chosen during request propagation.

Algorithm 1: Random Each outgoing arc has the same probability of being chosen during request propagation.

Algorithm 1: Random Each outgoing arc has the same probability of being chosen during request propagation.

Algorithm 1: Random

ALGORITHM 2: Greedy ORACLE

The Greedy ORACLE The oracle provides the sub-reviewer who will finish earlier.

The Greedy ORACLE The oracle provides the sub-reviewer who will finish earlier.

The Greedy ORACLE The oracle provides the sub-reviewer who will finish earlier.

ALGORITHM 3: FEEDBACK BASED

Probabilistic routing Arcs are labelled with rates to be used in probabilistic choices. 1 1+1 1 1 1 1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 1 1+1 1 1 1 1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 Can you review? * SE > 2 1 1+1 1 1 1 1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 Can you review? * SE > 2 NO 1 1+1 1 1 1 1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 Can you review? * SE > 2 NO 1 1+1 1 1 1 1-1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 1 Can you review? 1 1+1 1 1 1 1-1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 1 Can you review? Can you review? 1 1+1 1 1 1 1-1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 1 Can you review? YES Can you review? 1 1+1 1 1 1 1-1

Feedback-based rates Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 1 Can you review? YES Can you review? 1+1 1+1 1 1 1 1-1

Feedback-based rates

Feedback-based rates

ALGORITHM 4: con+dence-based

Feedback-based pheromones Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 FM: 4 FM: 3 SE: 5 SE: 1 Arcs labeled with one rate for each research topic.

Feedback-based pheromones Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 FM: 4 FM: 3 SE: 5 SE: 1 FM: 3 SE: 1 Arcs labeled with one rate for each research topic.

Feedback-based pheromones Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 FM: 0 SE: 0 FM: 4 FM: 3 SE: 5 SE: 1 FM: 3 SE: 1 Arcs labeled with one rate for each research topic.

Feedback-based pheromones Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 FM: 0 SE: 0 FM: 4 SE: 5 FM: 4 FM: 3 SE: 5 SE: 1 FM: 3 SE: 1 Arcs labeled with one rate for each research topic.

Feedback-based pheromones Paper XXX Due dd/mm/yy CoS: * FM > 3 * SE > 2 FM: 0 SE: 0 FM: 4 SE: 5 FM: 4 SE: 5 FM: 4 FM: 3 SE: 5 SE: 1 FM: 3 SE: 1 Arcs labeled with one rate for each research topic.

Confidence-based Rates

Confidence-based Rates

Confidence-based Rates

ewers e revi hoos c domly t) ran lmos (a

What's next? 1 Study the impact of the structure of the overlay network

What's next? 1 Study the impact of the structure of the overlay network 2 Study reputation-based strategies

What's next? 1 Study the impact of the structure of the overlay network 2 Study reputation-based strategies 3 Application to routing of messages in predicate-based communication (cf. SCEL)

Questions?

References Work-in-progress partially reported in: “A Computational Field Framework for Collaborative Task Execution in Volunteer Clouds”, Stefano Sebastio, Michele Amoretti and Alberto Lluch-Lafuente, draft [PDF] “Reputation-based Cooperation in the Clouds”, Alessandro Celestini, Alberto Lluch Lafuente, Philip Mayer, Stefano Sebastio, and Francesco Tiezzi, draft [PDF] See also: The science cloud platform. http://svn.pst.ifi.lmu.de/trac/scp/. P. Mayer et al. The Autonomic Cloud: A Vision of Voluntary, Peer-2-Peer Cloud Computing, 3rd Workshop on Challenges for Achieving Self- Awareness in Autonomic Systems, 2013.

Add a comment

Related presentations

Presentación que realice en el Evento Nacional de Gobierno Abierto, realizado los ...

In this presentation we will describe our experience developing with a highly dyna...

Presentation to the LITA Forum 7th November 2014 Albuquerque, NM

Un recorrido por los cambios que nos generará el wearabletech en el futuro

Um paralelo entre as novidades & mercado em Wearable Computing e Tecnologias Assis...

Microsoft finally joins the smartwatch and fitness tracker game by introducing the...

Related pages

Collaborative Task Execution In Volunteer Clouds (or how ...

Home; Technology; Collaborative Task Execution In Volunteer Clouds (or how to choose a sub-reviewer)
Read more

Kris Delaplain, Principal October 13, 2015. vol-un-teer n ...

... answers. Keep students on task ... Collaborative Task Execution In Volunteer Clouds (or how to choose a sub-reviewer) My talk at ...
Read more