icws09 lecue presentation
TRANSCRIPT
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Towards Scalability of Quality DrivenSemantic Web Service Composition
Freddy Lécué1 Nikolay Mehandjiev1
1The University of ManchesterBooth Street East, Manchester, UK
IEEE 7th International Conference on Web Services(ICWS 2009)
July 6th - 10th, 2009Los Angeles, CA, USA
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Outline
1 Introduction
2 Related Work
3 Background
4 Quality Model
5 A Scalable Approach
6 Experimentation
7 Conclusion
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Big Picture(Automation of) Web service composition in the Semantic Web.
Rather than Optimization (NP Hard!), we address ...Scalability of quality driven semantic service composition:
by selecting compositions based on:functional constraints;and QoS constraints.
s2
s1 s5
s3 s6
s7
T4
T2 T3 T6
T7
T8T1 T5
s4
s8
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Most of approaches focus on optimization wrt:QoS or functional parameters.L. Zeng, B. Benatallah et al.Quality Driven Web Services Composition.In WWW, pages 411–421, 2003.F. Lecue, A. Leger and A. DelteilOptimizing Causal Link Based Web Service Composition.In ECAI, pages 45–49, 2008.
Our work focuses on composition selection wrt:both latter criteria!
Selection
SelectionSemantics based
QoS baseds1 s5
s3 s6
s7
T4
T2 T3 T6
T7
T8T1 T5
s4
s8
s12, s
22, s
32, ...
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Semantic Web Service in a Nutshell
Parameters (i.e., Input and Output) of Web services insemantic Web are concepts referred to in an ontology T :
WSDL-S, SA-WSDL (W3C Proposed Recommendation);OWL-S profile level;WSMO capability level.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sx
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
Out_sy In_sx
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
Out_sy In_sx
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
Out_sy In_sx
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
NetWorkConnectionNetWorkConnection
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
NetWorkConnectionSlowNetWorkConnection
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
SlowNetWorkConnectionNetWorkConnection
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
VoIPIdIPAddress
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition and its Semantic Links
Semantic Link: Semantic connection between services;... more particulary between Output and Input parameters;... denoted by sly,x and valued by SimT (Out_sy , In_sx);
SimT is reduced to the five matchmaking functions[M.Paolucci et al. ISWC’02, Li and Horrocks WWW’03]:
Exact i.e., T |= Out_sy ≡ In_sx ;PlugIn i.e., T |= Out_sy v In_sx ;Subsume i.e., T |= In_sx v Out_sy ;Intersection i.e., T 6|= Out_sy u In_sx v ⊥;Disjoint i.e., T |= Out_sy u In_sx v ⊥;
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
NetWorkConnectionAddress
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Extra & Common Description in Semantic Links
Definition (Concept Abduction)Let L be a DL, Out_si , In_sj be two concepts in L, and T be aset of axioms in L. A Concept Abduction Problem (CAP),denoted as 〈L, Out_si , In_sj , T 〉 is finding a concept H ∈ Lsuch that T |= Out_si u H v In_sj .
The Extra Description H represents what is underspecified inOut_si in order to completely satisfy In_sj ;⇒ Explain why Out_si and In_sj can not be chained by a
robust semantic link.
The Common Description CD lcs(Out_si , In_sj) refers toinformation required by In_sj and effectively provided byOut_si .
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Extra & Common Description in Semantic Links
Definition (Concept Abduction)Let L be a DL, Out_si , In_sj be two concepts in L, and T be aset of axioms in L. A Concept Abduction Problem (CAP),denoted as 〈L, Out_si , In_sj , T 〉 is finding a concept H ∈ Lsuch that T |= Out_si u H v In_sj .
e.g., in case of a semantic link valued by the Subsumematching type.
Web service: sx
S Input Parameters
x
S Output Parameters
x
S Output Parameters
Web service: sy
S Input Parameters
y
y
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Extra & Common Description in Semantic Links
Definition (Concept Abduction)Let L be a DL, Out_si , In_sj be two concepts in L, and T be aset of axioms in L. A Concept Abduction Problem (CAP),denoted as 〈L, Out_si , In_sj , T 〉 is finding a concept H ∈ Lsuch that T |= Out_si u H v In_sj .
e.g., in case of a semantic link valued by the Subsumematching type.
Web service: sx
S Input Parameters
x
S Output Parameters
x
S Output Parameters
Web service: sy
S Input Parameters
y
y
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Composition Model
Process Model as a StatechartIts states refer to services;Its transitions are labelled with semantic links;with basic composition constructs such as Sequence,conditional branching (i.e., OR-Branching), concurrentthreads (i.e., AND-Branching).
Legend
Connection
Slow
Output Parameter
Input Parameter
T: Task
s: Service
Semantic Link sl
Network
Connection
Network
s1 s5
s2 s3
ANDBranching
s6
s7
sl15,7
sl12,3
sl11,4
sl15,6
sl14,5
sl16,8
T4
T2 T3 T6
T7
T8T1 T5
sl11,2 sl13,5
s4
s8
OR-Branching
Sequence
sl17,8
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Semantic Links & Services
q(sli,j) for Elementary Semantic Links sli,jCommon Description rate qcd ∈ (0, 1]:
qcd(sli,j) =|lcs(Out_si , In_sj)|
|H∈〈L,Out_si ,In_sj ,T 〉| + |lcs(Out_si , In_sj)|
Matching Quality qm ∈ (0, 1], valued by SimT (Out_si , In_sj)(Exact: 1, PlugIn: 3
4 , Subsume: 12 , Intersection: 1
4 ).
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Semantic Links & Services
q(sli,j) for Elementary Semantic Links sli,jCommon Description rate qcd ∈ (0, 1]:
qcd(sli,j) =|lcs(Out_si , In_sj)|
|H∈〈L,Out_si ,In_sj ,T 〉| + |lcs(Out_si , In_sj)|
Matching Quality qm ∈ (0, 1], valued by SimT (Out_si , In_sj)(Exact: 1, PlugIn: 3
4 , Subsume: 12 , Intersection: 1
4 ).
Web service: sxSemantic connection:
S Input Parameters
y
S Input Parameters
xS Output Parameters
y
S Output Parameters
x
Web service: sy
SimT
SlowNetWorkConnectionNetWorkConnection
(Subsume i.e., )12_
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Semantic Links & Services
q(sli,j) for Elementary Semantic Links sli,jCommon Description rate qcd ∈ (0, 1]:
qcd(sli,j) =|lcs(Out_si , In_sj)|
|H∈〈L,Out_si ,In_sj ,T 〉| + |lcs(Out_si , In_sj)|
Matching Quality qm ∈ (0, 1], valued by SimT (Out_si , In_sj)(Exact: 1, PlugIn: 3
4 , Subsume: 12 , Intersection: 1
4 ).
q(si) for Elementary Services si
Execution Price qpr ∈ <+;Response Time qt ∈ <+.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Semantic Links & Services
q(sli,j) for Elementary Semantic Links sli,jCommon Description rate qcd ∈ (0, 1]:
qcd(sli,j) =|lcs(Out_si , In_sj)|
|H∈〈L,Out_si ,In_sj ,T 〉| + |lcs(Out_si , In_sj)|
Matching Quality qm ∈ (0, 1], valued by SimT (Out_si , In_sj)(Exact: 1, PlugIn: 3
4 , Subsume: 12 , Intersection: 1
4 ).
q(si) for Elementary Services si
Execution Price qpr ∈ <+;Response Time qt ∈ <+.
QoS-extended quality vector of a semantic link sli,j∗q (sli,j)
.= (q(si), q(sli,j), q(sj))
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Composition
Quality Aggregation Rules for Compositions
Composition Quality Criterion
Construct Semantic Non FunctionalQcd Qm Qt Qpr
Sequential/ 1|sl|
∑sl qcd(sl)
∏sl qm(sl)
∑s qt(s) ∑
s qpr (s)AND- Branching maxs qt(s)
OR-Branching∑
sl qcd(sl).psl∑
sl qm(sl).psl∑
s qt(s).ps∑
s qpr (s).ps
Legend
Connection
Slow
Output Parameter
Input Parameter
T: Task
s: Service
Semantic Link sl
Network
Connection
Network
s1 s5
s2 s3
ANDBranching
s6
s7
sl15,7
sl12,3
sl11,4
sl15,6
sl14,5
sl16,8
T4
T2 T3 T6
T7
T8T1 T5
sl11,2 sl13,5
s4
s8
OR-Branching
Sequence
sl17,8
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Composition
Quality Aggregation Rules for Compositions
Composition Quality Criterion
Construct Semantic Non FunctionalQcd Qm Qt Qpr
Sequential/ 1|sl|
∑sl qcd(sl)
∏sl qm(sl)
∑s qt(s) ∑
s qpr (s)AND- Branching maxs qt(s)
OR-Branching∑
sl qcd(sl).psl∑
sl qm(sl).psl∑
s qt(s).ps∑
s qpr (s).ps
Connection
Slow
Output Parameter
Input Parameter
T: Task
s: Service
Network
Connection
Legend
Semantic Link slNetwork sl16,8
s1 s5
s2 s3 s6sl12,3 sl15,6
T2 T3 T6
T8T1 T5
sl11,2 sl13,5
s8
Sequence
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Composition
Quality Aggregation Rules for Compositions
Composition Quality Criterion
Construct Semantic Non FunctionalQcd Qm Qt Qpr
Sequential/ 1|sl|
∑sl qcd(sl)
∏sl qm(sl)
∑s qt(s) ∑
s qpr (s)AND- Branching maxs qt(s)
OR-Branching∑
sl qcd(sl).psl∑
sl qm(sl).psl∑
s qt(s).ps∑
s qpr (s).ps
Output Parameter
Input Parameter
T: Task
s: Service
Legend
Semantic Link sl
sl15,7
s5AND
Branching
s6
s7
sl15,6
T6
T7
T5
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Composition
Quality Aggregation Rules for Compositions
Composition Quality Criterion
Construct Semantic Non FunctionalQcd Qm Qt Qpr
Sequential/ 1|sl|
∑sl qcd(sl)
∏sl qm(sl)
∑s qt(s) ∑
s qpr (s)AND- Branching maxs qt(s)
OR-Branching∑
sl qcd(sl).psl∑
sl qm(sl).psl∑
s qt(s).ps∑
s qpr (s).ps
Semantic Link sl
Connection
Slow
Output Parameter
Input Parameter
T: Task
s: Service
Network
Connection
Legend
Network
s1
s2
sl11,4T4
T2
T1
sl11,2
s4
OR-Branching
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality Criteria for Composition
Quality Aggregation Rules for Compositions
Composition Quality Criterion
Construct Semantic Non FunctionalQcd Qm Qt Qpr
Sequential/ 1|sl|
∑sl qcd(sl)
∏sl qm(sl)
∑s qt(s) ∑
s qpr (s)AND- Branching maxs qt(s)
OR-Branching∑
sl qcd(sl).psl∑
sl qm(sl).psl∑
s qt(s).ps∑
s qpr (s).ps
A Quality Vector for Web Service Composition“A” way to differentiate compositions:
Q(c).= (Qcd(c), Qm(c), Qt(c), Qpr (c))
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Web Service Composition Driven CSP
CSP FormalizationFormalization as a triple (T , D, C):
T is the set of tasks (variables) {T1, T2, ..., Tn};D is the set of domains {D1, D2, ..., Dn} i.e., services;C is the set of constraints i.e., local CL and global CG.
e.g.,1|slA
i,j |∑slAi,j
qcd(slAi,j) ≥ v , v ∈ [0, 1]
∑Ti
qpr (Ti) ≤ v , v ∈ <+
Main Goal to AchieveAn assignment (si , Ti)1≤i≤n i.e., (service, task)
with si,1≤i≤n ∈ Di,1≤i≤n;which satisfies all the constraints C.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
A Stochastic Search Method (1)
PrinciplesSacrificing completness (i.e., all solutions) for speed;Based on a simple idea: computing “a single” solution.
Our ApproachAdaptation of the Hill Climbing algorithm.→ Appropriate for a large number of services.
S. Russell and P. Norvig.Artificial Intelligence: A Modern Approach.Ed. Prentice-Hall, 1995.
Computational ComplexityCSP based search methods: Exponential!Stochastic search methods (e.g., Hill Climbing) scale better!
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
A Stochastic Search Method (2)
RequirementsAn evaluation function f for each composition c:
f (c) =ωcd Q̂cd(c) + ωmQ̂m(c)
ωpr Q̂pr (c) + ωtQ̂t(c)
An adjacency function: c1 and c2 are adjacent to eachother if they differ in exactly one assignment (s, T ).
Algorithm in Details1) Let’s start with a random composition cfinal .2) f -Evaluation of all ci,1≤i≤n adjacent to cfinal .
If ∃i such that f (cfinal) ≤ f (ci) then f (cfinal)← f (ci).
3) Iteration until all constraints are satisfied by cfinal .4) If no solution, constraints relaxing.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Evolution of Constraints SatisfactionThe more tasks, services the more time consuming!
Evolution of Composition QualityOptimal composition: High Time consuming!Compositions that satisfy constraints: More scalable!
Search Process vs. DL Reasoning (|T | > 100, |s| > 350)
DL reasoning is the most time consuming process!Large number of potential semantic links.Critical complexity of DL abduction.
Vs. State-of-the-art Approaches (T = 300 |s| > 280)
Adoption of stochastic search method for large domains!No exponential search required.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Evolution of Constraints SatisfactionThe more tasks, services the more time consuming!
Evolution of Composition QualityOptimal composition: High Time consuming!Compositions that satisfy constraints: More scalable!
Search Process vs. DL Reasoning (|T | > 100, |s| > 350)
DL reasoning is the most time consuming process!Large number of potential semantic links.Critical complexity of DL abduction.
Vs. State-of-the-art Approaches (T = 300 |s| > 280)
Adoption of stochastic search method for large domains!No exponential search required.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Evolution of Constraints SatisfactionThe more tasks, services the more time consuming!
Evolution of Composition QualityOptimal composition: High Time consuming!Compositions that satisfy constraints: More scalable!
Search Process vs. DL Reasoning (|T | > 100, |s| > 350)
DL reasoning is the most time consuming process!Large number of potential semantic links.Critical complexity of DL abduction.
Vs. State-of-the-art Approaches (T = 300 |s| > 280)
Adoption of stochastic search method for large domains!No exponential search required.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Evolution of Constraints SatisfactionThe more tasks, services the more time consuming!
Evolution of Composition QualityOptimal composition: High Time consuming!Compositions that satisfy constraints: More scalable!
Search Process vs. DL Reasoning (|T | > 100, |s| > 350)
DL reasoning is the most time consuming process!Large number of potential semantic links.Critical complexity of DL abduction.
Vs. State-of-the-art Approaches (T = 300 |s| > 280)
Adoption of stochastic search method for large domains!No exponential search required.
Introduction Related Work Background Quality Model A Scalable Approach Experimentation Conclusion
Quality-driven Semantic Web Service Composition[Theoretical]
A general and extensible model to evaluate compositions;Scalabilty: A solution rather than the most optimal.
CSP formalization.Adaptation of a stochastic search method.
[Experimental]Good computation costs despite the off-line DL reasoning.
Future WorkExtension with more composition constructs;Considering a finer abduction operator;Dynamic Distribution of the CSP on different peers;Focusing on a process that reduces the number ofsemantic links: Macro composition of Web services.