際際滷

際際滷Share a Scribd company logo
Incentivising Resource Sharing
in Federated Clouds
Eduardo Falc達o
eduardolfalcao@lsd.ufcg.edu.br
Francisco Brasileiro
Andrey Brito
Jos辿 Luis Vivas
Challenge
Promotion of cooperation among selfish
individuals in a decentralized context;
Incentive mechanisms to prevent
collaborative peers from defecting from the
federation;
Peers need to take efficient decisions
To whom should I donate?
How much should I donate?
2
Baseline  Network of Favors
Network of Favors (NoF)
each peer keeps a local record of the past
interactions with other peers.
Peer Provided
Favors
Consumed
Favors
Credit
D 300 150 150
B 100 55 45
E 30 5 25
C 15 30 0
F 0 200 0
A
{ 3
Problem Statement
Whenever there is contention of resources,
the collaborative actors are prioritized;
NoF works fine in these scenarios.
Peer Provided
Favors
Consumed
Favors
Credit
D 300 150 150
B 100 55 45
E 30 5 25
C 15 30 0
F 0 200 0
A
{
requests
+10
+10
+10
+10
+10
free: +20
4
Problem Statement
When there isnt contention of resources, some
free riders are able to consume the exceeding
resources.
These resources wont be reciprocated in the future.
Peer Provided
Favors
Consumed
Favors
Credit
D 300 150 150
B 100 55 45
E 30 5 25
C 15 30 0
F 0 200 0
A
{
requests
0
+10
free: +20
0
0
0
5
Solution
$$ =

$
 =

p
 Satisfaction-Driven NoF (SD-NoF)
 Context: P2P opportunistic desktop
grids;
 Low cost for maintenance and
hosting;
 Provides all exceeding
resources to maximize the
satisfaction of the peers.
 Fairness-Driven NoF (FD-NoF)
 Context: P2P federation of
private cloud providers;
 High cost due to purchase,
maintenance, energy, hosting,
etc;
 Regulates the amount of
provided resources in order to
guarantee good levels of
fairness
probability of being answered level of reciprocity
6
Simulation Model
: number of peers;
: % free riders;
The simulation poceeds in steps;
Collaborators and free riders have the same total
resource capacity (), and the same amount of
demand (  );
SD-NoF  provides ;
FD-NoF  provides 0,  ;
: probability of consuming;
Colaborators  [consuming state, provider state];
Free riders  [consuming state], critical case,  = 1.
factors
7
Scenarios
Designed to assess the different scenarios of
contention of resources between
collaborators ():
 = 0.5  low resource contention
 = 1  moderate resource contention
 {2, 4}  high resource contention
requested
provided
 =
10
20
 =
10
10
  =
10
5
factors
Constant  n=100; f=75%; steps=10000; C=1
Variable   and 
8
SD-NoF
Proposal: FD-NoF
collaborators should be
equipped with a control
mechanism that enables them
to regulate the amount of
resources donated
9
Peer Provided
Favors
Consumed
Favors
Credit
D 300 150 150
B 100 55 45
E 30 5 25
C 15 30 0
F 0 200 0
Peer Provided
Favors
Consumed
Favors
Credit Current
Fairness
Previous
Fairness
D 300 150 150 2 ...
B 100 55 45 1.81 ...
E 30 5 25 6 ...
C 15 30 0 0.5 ...
F 0 200 0 0 ...
FD-NoF
NoF + Feedback Control Loop  FD-NoF
A
{
10
FD-NoF
= %  
A
E
D
C
B
Max(C)
80%
100%
65%
50%
11
FD-NoF: satisfaction
Resource contention=0.5
Average deterioration: 4%
12
FD-NoF: fairness
Resource contention=0.5
Average improvement: 44%
Is it worth sacrificing 4% of
satisfaction to increase 44% of
fairness?
13
Concluding Remarks
FD-NoF shows good evidences that achieves
its goal:
Low resource contention
 Satisfaction  -4%
Fairness  +44%
Moderate and high resource contention: similar
results to SD-NoF.
14
Future Work
Improve the Feedback Control Loop;
Validation of the results in a realistic scenario;
Fogbow middleware
http://www.fogbowcloud.org/
EUBrazilCC project: federation of private cloud
providers
http://eubrazilcloudconnect.eu/
15
Thank you!
Questions?
eduardolfalcao@lsd.ufcg.edu.br
16

More Related Content

More from Eduardo de Lucena Falc達o (17)

Aula 3 - Introdu巽達o a cloud computingAula 3 - Introdu巽達o a cloud computing
Aula 3 - Introdu巽達o a cloud computing
Eduardo de Lucena Falc達o
Aula 2   introdu巽達o a sistemas distribu鱈dosAula 2   introdu巽達o a sistemas distribu鱈dos
Aula 2 introdu巽達o a sistemas distribu鱈dos
Eduardo de Lucena Falc達o
Aula 1 - Estudando o problema a ser resolvidoAula 1 - Estudando o problema a ser resolvido
Aula 1 - Estudando o problema a ser resolvido
Eduardo de Lucena Falc達o
Apresenta巽達o do CursoApresenta巽達o do Curso
Apresenta巽達o do Curso
Eduardo de Lucena Falc達o
II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...
II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...
Eduardo de Lucena Falc達o
INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...
INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...
Eduardo de Lucena Falc達o
Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...
Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...
Eduardo de Lucena Falc達o
SEAC: Um Simulador Online para Ensino de Arquitetura de ComputadoresSEAC: Um Simulador Online para Ensino de Arquitetura de Computadores
SEAC: Um Simulador Online para Ensino de Arquitetura de Computadores
Eduardo de Lucena Falc達o
Processo de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRASProcesso de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRAS
Processo de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRAS
Eduardo de Lucena Falc達o
Desafios da tecnologia da computa巽達o e ind炭striaDesafios da tecnologia da computa巽達o e ind炭stria
Desafios da tecnologia da computa巽達o e ind炭stria
Eduardo de Lucena Falc達o
TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I
TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I
Eduardo de Lucena Falc達o
Tutorial Shell ScriptTutorial Shell Script
Tutorial Shell Script
Eduardo de Lucena Falc達o
E-InfoE-Info
E-Info
Eduardo de Lucena Falc達o
Descoberta de Conhecimento em Bancos de Dados e Minera巽達o de DadosDescoberta de Conhecimento em Bancos de Dados e Minera巽達o de Dados
Descoberta de Conhecimento em Bancos de Dados e Minera巽達o de Dados
Eduardo de Lucena Falc達o
Ambiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de ProcessadoresAmbiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de Processadores
Ambiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de Processadores
Eduardo de Lucena Falc達o
Introdu巽達o pr叩tica ao X3DIntrodu巽達o pr叩tica ao X3D
Introdu巽達o pr叩tica ao X3D
Eduardo de Lucena Falc達o
Lab teve2011.1Lab teve2011.1
Lab teve2011.1
Eduardo de Lucena Falc達o
Aula 3 - Introdu巽達o a cloud computingAula 3 - Introdu巽達o a cloud computing
Aula 3 - Introdu巽達o a cloud computing
Eduardo de Lucena Falc達o
Aula 2   introdu巽達o a sistemas distribu鱈dosAula 2   introdu巽達o a sistemas distribu鱈dos
Aula 2 introdu巽達o a sistemas distribu鱈dos
Eduardo de Lucena Falc達o
Aula 1 - Estudando o problema a ser resolvidoAula 1 - Estudando o problema a ser resolvido
Aula 1 - Estudando o problema a ser resolvido
Eduardo de Lucena Falc達o
Apresenta巽達o do CursoApresenta巽達o do Curso
Apresenta巽達o do Curso
Eduardo de Lucena Falc達o
II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...
II EPI - Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cad...
Eduardo de Lucena Falc達o
INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...
INFORMAO ARQUEOLGICA DE ING: preserva巽達o, acesso e uso a partir de um amb...
Eduardo de Lucena Falc達o
Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...
Estudo e Avalia巽達o do Problema de Otimiza巽達o da Multiplica巽達o de Cadeias de M...
Eduardo de Lucena Falc達o
SEAC: Um Simulador Online para Ensino de Arquitetura de ComputadoresSEAC: Um Simulador Online para Ensino de Arquitetura de Computadores
SEAC: Um Simulador Online para Ensino de Arquitetura de Computadores
Eduardo de Lucena Falc達o
Processo de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRASProcesso de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRAS
Processo de Tradu巽達o Autom叩tica da L鱈ngua Portuguesa para LIBRAS
Eduardo de Lucena Falc達o
Desafios da tecnologia da computa巽達o e ind炭striaDesafios da tecnologia da computa巽達o e ind炭stria
Desafios da tecnologia da computa巽達o e ind炭stria
Eduardo de Lucena Falc達o
TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I
TCC - PUBLICAO E ACESSO A CONTEDOS 3D ATRAVS DA WEB: O CASO DO MUSEU3I
Eduardo de Lucena Falc達o
Tutorial Shell ScriptTutorial Shell Script
Tutorial Shell Script
Eduardo de Lucena Falc達o
Descoberta de Conhecimento em Bancos de Dados e Minera巽達o de DadosDescoberta de Conhecimento em Bancos de Dados e Minera巽達o de Dados
Descoberta de Conhecimento em Bancos de Dados e Minera巽達o de Dados
Eduardo de Lucena Falc達o
Ambiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de ProcessadoresAmbiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de Processadores
Ambiente de Simula巽達o Gr叩fica 3D para Ensino da Arquitetura de Processadores
Eduardo de Lucena Falc達o
Introdu巽達o pr叩tica ao X3DIntrodu巽達o pr叩tica ao X3D
Introdu巽達o pr叩tica ao X3D
Eduardo de Lucena Falc達o

Recently uploaded (20)

The Broccoli Dog's inner voice (look A)
The Broccoli Dog's inner voice  (look A)The Broccoli Dog's inner voice  (look A)
The Broccoli Dog's inner voice (look A)
merasan
How to Modify Existing Web Pages in Odoo 18
How to Modify Existing Web Pages in Odoo 18How to Modify Existing Web Pages in Odoo 18
How to Modify Existing Web Pages in Odoo 18
Celine George
What do students really understand about academic integrity? International Ce...
What do students really understand about academic integrity? International Ce...What do students really understand about academic integrity? International Ce...
What do students really understand about academic integrity? International Ce...
Thomas Lancaster
Kaun TALHA quiz Finals -- El Dorado 2025
Kaun TALHA quiz Finals -- El Dorado 2025Kaun TALHA quiz Finals -- El Dorado 2025
Kaun TALHA quiz Finals -- El Dorado 2025
Conquiztadors- the Quiz Society of Sri Venkateswara College
Database population in Odoo 18 - Odoo slides
Database population in Odoo 18 - Odoo slidesDatabase population in Odoo 18 - Odoo slides
Database population in Odoo 18 - Odoo slides
Celine George
Computer Network Unit IV - Lecture Notes - Network Layer
Computer Network Unit IV - Lecture Notes - Network LayerComputer Network Unit IV - Lecture Notes - Network Layer
Computer Network Unit IV - Lecture Notes - Network Layer
Murugan146644
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQCShibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Rudra Chakraborty
Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1...
Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1...Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1...
Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1...
pinkdvil200
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa PagpapahalagaVE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
chezeltaylan1
Adventure Activities Final By H R Gohil Sir
Adventure Activities Final By H R Gohil SirAdventure Activities Final By H R Gohil Sir
Adventure Activities Final By H R Gohil Sir
GUJARATCOMMERCECOLLE
The Story Behind the Abney Park Restoration Project by Tom Walker
The Story Behind the Abney Park Restoration Project by Tom WalkerThe Story Behind the Abney Park Restoration Project by Tom Walker
The Story Behind the Abney Park Restoration Project by Tom Walker
History of Stoke Newington
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptxTLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
RizaBedayo
How to attach file using upload button Odoo 18
How to attach file using upload button Odoo 18How to attach file using upload button Odoo 18
How to attach file using upload button Odoo 18
Celine George
Integrative Creative Works of Selected Philippines and SEA arts.pptx
Integrative Creative Works of Selected Philippines and SEA arts.pptxIntegrative Creative Works of Selected Philippines and SEA arts.pptx
Integrative Creative Works of Selected Philippines and SEA arts.pptx
BeauDarrylPanizales
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptxPrimary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Ms. Pooja Bhandare
Unit 3_Hash function and MD5 working.pdf
Unit 3_Hash function and MD5 working.pdfUnit 3_Hash function and MD5 working.pdf
Unit 3_Hash function and MD5 working.pdf
KanchanPatil34
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Sean Fisher
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
PRADEEP ABOTHU
Summary of the 2024 Western CMV Safety Summit
Summary of the 2024 Western CMV Safety SummitSummary of the 2024 Western CMV Safety Summit
Summary of the 2024 Western CMV Safety Summit
UGPTI
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
History of Stoke Newington
The Broccoli Dog's inner voice (look A)
The Broccoli Dog's inner voice  (look A)The Broccoli Dog's inner voice  (look A)
The Broccoli Dog's inner voice (look A)
merasan
How to Modify Existing Web Pages in Odoo 18
How to Modify Existing Web Pages in Odoo 18How to Modify Existing Web Pages in Odoo 18
How to Modify Existing Web Pages in Odoo 18
Celine George
What do students really understand about academic integrity? International Ce...
What do students really understand about academic integrity? International Ce...What do students really understand about academic integrity? International Ce...
What do students really understand about academic integrity? International Ce...
Thomas Lancaster
Database population in Odoo 18 - Odoo slides
Database population in Odoo 18 - Odoo slidesDatabase population in Odoo 18 - Odoo slides
Database population in Odoo 18 - Odoo slides
Celine George
Computer Network Unit IV - Lecture Notes - Network Layer
Computer Network Unit IV - Lecture Notes - Network LayerComputer Network Unit IV - Lecture Notes - Network Layer
Computer Network Unit IV - Lecture Notes - Network Layer
Murugan146644
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQCShibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Shibpur_ IIEST_Preliminary Round Impetus Quiz Fest by Rudra Chakraborty BQC
Rudra Chakraborty
Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1...
Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1...Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1 2024  Lesson Plan M1...
Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1 2024 Lesson Plan M1...
pinkdvil200
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa PagpapahalagaVE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
VE7_Q4W4D2.pptxBaitang 7 Edukasyon sa Pagpapahalaga
chezeltaylan1
Adventure Activities Final By H R Gohil Sir
Adventure Activities Final By H R Gohil SirAdventure Activities Final By H R Gohil Sir
Adventure Activities Final By H R Gohil Sir
GUJARATCOMMERCECOLLE
The Story Behind the Abney Park Restoration Project by Tom Walker
The Story Behind the Abney Park Restoration Project by Tom WalkerThe Story Behind the Abney Park Restoration Project by Tom Walker
The Story Behind the Abney Park Restoration Project by Tom Walker
History of Stoke Newington
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptxTLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
TLE 7 - 2nd Topic - Codes and Standards in Industrial Arts Services.pptx
RizaBedayo
How to attach file using upload button Odoo 18
How to attach file using upload button Odoo 18How to attach file using upload button Odoo 18
How to attach file using upload button Odoo 18
Celine George
Integrative Creative Works of Selected Philippines and SEA arts.pptx
Integrative Creative Works of Selected Philippines and SEA arts.pptxIntegrative Creative Works of Selected Philippines and SEA arts.pptx
Integrative Creative Works of Selected Philippines and SEA arts.pptx
BeauDarrylPanizales
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptxPrimary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Primary metabolites(Carbohydrates Acacia, Agar, Tragacanth, Honey).pptx
Ms. Pooja Bhandare
Unit 3_Hash function and MD5 working.pdf
Unit 3_Hash function and MD5 working.pdfUnit 3_Hash function and MD5 working.pdf
Unit 3_Hash function and MD5 working.pdf
KanchanPatil34
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Fillable Strategies for Understanding and Supporting Neurodivergent Multiling...
Sean Fisher
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
THE NEEDS OF NORMAL CHILDREN THROUGH THE STAGES OF DEVELOPMENTAL AND PARENTAL...
PRADEEP ABOTHU
Summary of the 2024 Western CMV Safety Summit
Summary of the 2024 Western CMV Safety SummitSummary of the 2024 Western CMV Safety Summit
Summary of the 2024 Western CMV Safety Summit
UGPTI
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
South Hornsey: The Lost Local Authority that Merged with Stoke Newington by T...
History of Stoke Newington

Dais 2015

  • 1. Incentivising Resource Sharing in Federated Clouds Eduardo Falc達o eduardolfalcao@lsd.ufcg.edu.br Francisco Brasileiro Andrey Brito Jos辿 Luis Vivas
  • 2. Challenge Promotion of cooperation among selfish individuals in a decentralized context; Incentive mechanisms to prevent collaborative peers from defecting from the federation; Peers need to take efficient decisions To whom should I donate? How much should I donate? 2
  • 3. Baseline Network of Favors Network of Favors (NoF) each peer keeps a local record of the past interactions with other peers. Peer Provided Favors Consumed Favors Credit D 300 150 150 B 100 55 45 E 30 5 25 C 15 30 0 F 0 200 0 A { 3
  • 4. Problem Statement Whenever there is contention of resources, the collaborative actors are prioritized; NoF works fine in these scenarios. Peer Provided Favors Consumed Favors Credit D 300 150 150 B 100 55 45 E 30 5 25 C 15 30 0 F 0 200 0 A { requests +10 +10 +10 +10 +10 free: +20 4
  • 5. Problem Statement When there isnt contention of resources, some free riders are able to consume the exceeding resources. These resources wont be reciprocated in the future. Peer Provided Favors Consumed Favors Credit D 300 150 150 B 100 55 45 E 30 5 25 C 15 30 0 F 0 200 0 A { requests 0 +10 free: +20 0 0 0 5
  • 6. Solution $$ = $ = p Satisfaction-Driven NoF (SD-NoF) Context: P2P opportunistic desktop grids; Low cost for maintenance and hosting; Provides all exceeding resources to maximize the satisfaction of the peers. Fairness-Driven NoF (FD-NoF) Context: P2P federation of private cloud providers; High cost due to purchase, maintenance, energy, hosting, etc; Regulates the amount of provided resources in order to guarantee good levels of fairness probability of being answered level of reciprocity 6
  • 7. Simulation Model : number of peers; : % free riders; The simulation poceeds in steps; Collaborators and free riders have the same total resource capacity (), and the same amount of demand ( ); SD-NoF provides ; FD-NoF provides 0, ; : probability of consuming; Colaborators [consuming state, provider state]; Free riders [consuming state], critical case, = 1. factors 7
  • 8. Scenarios Designed to assess the different scenarios of contention of resources between collaborators (): = 0.5 low resource contention = 1 moderate resource contention {2, 4} high resource contention requested provided = 10 20 = 10 10 = 10 5 factors Constant n=100; f=75%; steps=10000; C=1 Variable and 8
  • 9. SD-NoF Proposal: FD-NoF collaborators should be equipped with a control mechanism that enables them to regulate the amount of resources donated 9
  • 10. Peer Provided Favors Consumed Favors Credit D 300 150 150 B 100 55 45 E 30 5 25 C 15 30 0 F 0 200 0 Peer Provided Favors Consumed Favors Credit Current Fairness Previous Fairness D 300 150 150 2 ... B 100 55 45 1.81 ... E 30 5 25 6 ... C 15 30 0 0.5 ... F 0 200 0 0 ... FD-NoF NoF + Feedback Control Loop FD-NoF A { 10
  • 11. FD-NoF = % A E D C B Max(C) 80% 100% 65% 50% 11
  • 13. FD-NoF: fairness Resource contention=0.5 Average improvement: 44% Is it worth sacrificing 4% of satisfaction to increase 44% of fairness? 13
  • 14. Concluding Remarks FD-NoF shows good evidences that achieves its goal: Low resource contention Satisfaction -4% Fairness +44% Moderate and high resource contention: similar results to SD-NoF. 14
  • 15. Future Work Improve the Feedback Control Loop; Validation of the results in a realistic scenario; Fogbow middleware http://www.fogbowcloud.org/ EUBrazilCC project: federation of private cloud providers http://eubrazilcloudconnect.eu/ 15

Editor's Notes

  • #3: Once the federation we are dealing is completely decentralized one of the greatest challenges we face is the promotion of cooperation among the participants. In P2P networks, in a first glance, peers dont know each other and are assumed to be selfish and to have economic incentive to become free riders, which are the non-reciprocators peers. Therefore, some type of incentive mechanism must exist in order to maximize the benefits of the peers staying in that federation, otherwise, they would leave it. A federation will provide good results to a peer only if he takes efficient decisions, which includes not donating to free riders. Thus, our incentive mechanism should help the peer to answer the following questions that they naturally have: To whom should I donate? (here, the mechanism should point out collaborative partners) How much should I donate? (once I do not trust anybody, should I donate all my exceeding resources to this unknown peer, or should I increase this ammount gradually?)
  • #4: Network of Favors (NoF): Ensures the prioritization of favors to colaborators ; In NoF, each peer keeps a local record of his past interactions with other peers. For example: peer D provided 300 units of favor to peer A, and condumed 150 units of favor from peer A, which gives him a credit of 150 units of favor with peer A. A unit of favor could be a measure that involves a processing power (which could be described as a flavour of the instance) multiplied by the time it was lent. As you can note, this table is ordered by the credit, and whenever peer A has exceeding resources, he will always prioritize the peers in the top of this table. This is how the NoF reward colaborative peers.
  • #5: Whenever there is contention of resources the Network of Favors works fine: the collaborative actors are prioritized and granted favors, and the free riders are left with a low degree of success. Thats what is happening in this case. The peer A has 20 units of free resources, and each peer in the table is requesting 10 units of favor. A will grant D 10 units, and the last 10 units will be provided to B.
  • #6: But in scenarios of low contention of resources, the surplus resources of the federation can be consumed by the free riders. In this new example, only peer F is requesting resources. Note that, for this single case, there is plenty of resources in the federation, and this increases the probability of a free rider grant more favors. We might note that peer F seems to be a free rider, because he didnt provided any favor to A, he just consumed favors. In this scenario (low contention of resources) the peer F would grant this favor, and if he really is a free rider, this resource will not be reciprocated to A in the future.
  • #7: In order to better quantify the performance of the peers in different scenarios of contention of resources, we created these two metrics: satisfaction and fairness. Satisfaction is measured by the amount of resources consumed over the amount of resources requested, which could be also referred as the probability of being answered when requesting resources. For example, if a peer requests 10 resources and he is granted these 10 resources, then, he has the best level of satisfaction possible, which is 1. Fairnes is the amount of consumed resources over the amount of donated resources. It could also be referred as the degree of reciprocity. For example, if a peer provides 10 resources and consumes only 1 unit of resources, he has a low level of fairness, 0.1. If he consumed the 10 resources it had provided, then he would have a good level of fairness, 1. Based on these 2 metrics we took the liberty of renaming the NoF, which was previously proposed to opportunistic desktop grids, to Satisfaction-Driven NoF. But why? In this NoF, every peer always provides all of its free resources to try to maximize your own satisfaction. The idea is that when he provides more resources, higher is the probability of another peer consume them. Then, the providing peer would increase his credit with the consuming peer, and this would get him to the top of the other peers table. This would increase the probability of this peer be answered when he need resources. In this case, this could be done due to the low cost of resources. But, when we come to a scenario that involves high costs in the business model, like federation of cloud providers, the costs should not be neglected. Then, in these scenarios, the fairness is the metric to be maximized, since it involves the donation of resources. Basically, what we propose is that the peers should regulate the amount of provided resources, so that the federation could avoid scenarios of very low contention of resources, which is good for free riders.
  • #8: Now that we have these 2 metrics, fairness and satisfaction, we want to understand how peers perform in SD-NoF and in our proposed FD-NoF. Our simulation have the following factors, which are in red. Each simulation of a p2p network is composed by: n: the number of peers f: the amount, in percentage, of free riders steps: the number of steps C: the amount of resources each peer has D: the demand of each peer PI: the probability of being in consuming state One simplification that we did is about the time. Here it is not continuous, but discrete. In our first evaluation, we wanted to know how the set of collaborators perform in relation to free riders. Thus, all collaborators have the same Capacitys and Demands. Further, we eliminate this simplification to better understand how collaborators with higher Demand perform in comparison to another with lower demand. In the SD-NoF, a collaborator always supplies all of its free resources, in order to try to maximize his Satisfaction.
  • #9: We designed the scenarios in terms of the contention of resources. What we want here is to better understand how peers perform in different levels of contention. In mathematical terms, the contention of resources (kappa) is measured by the amount of requested resources by the set of collaborators in consuming state divided by the amount of resources supplied by the set of collaborators in provider state. For example, if the consuming collaborators request in total 10 resources. If there were a set of collaborator that together could donate 20 resources, then we would have a low resource contention scenario. Note that there would be 10 exceeding resources that could be consumed by free riders. Our proposal is that the set of collaborator could regulate the amount of supplied resources to the federeation and therefore control the level of contention. Note that this set of collaborators could reduce the total amount of resources supplied from 20 to 10, and so they would have a a moderate contention of resources. Nota that here, there arent surplus resources that the free riders could allocate. And in a similar way they could increase even more the level of contention, what would give us a scenario of high level of contention.
  • #10: We advocate that collaborators should be equipped with a control mechanism that enables them to regulate the amount of resources donated, thereby impacting the level of resource contention in the system. This should be done in such a way that high values for fairness are attained without impairing that of satisfaction.
  • #14: Ao analisarmos K=0.5 e K=1, os casos onde a justi巽a pode ser afetada devido sobra de recursos, podemos ver que os colaboradores apresentaram uma melhoria em ambos os cen叩rios. Para K=0.5, os colaboradores tiveram ganho absoluto m辿dio de 20% de justi巽a e e ganho relativo de 38%. Ent達o a pergunta relativa ao trade-off seria: vale a pena trocar 1% de satisfa巽達o por 38% de justi巽a?