1 © Copyright 2010 Dieter Fensel and Ioan Toma Semantic Web Services Service Science.
1 © Copyright 2010 Dieter Fensel and Ioan Toma Semantic Web Services Applications.
-
Upload
lizbeth-quinn -
Category
Documents
-
view
213 -
download
0
Transcript of 1 © Copyright 2010 Dieter Fensel and Ioan Toma Semantic Web Services Applications.
1© Copyright 2010 Dieter Fensel and Ioan Toma
Semantic Web Services
Applications
2
Where are we?
# Title
1 Introduction
2 Web Science
3 Service Science
4 Web services
5 Web2.0 services
6 Semantic Web
7 Web Service Modeling Ontology (WSMO)
8 Web Service Modeling Language (WSML)
9 Web Service Execution Environment (WSMX)
10 OWL-S and others
11 Light-weight Annotations
12 Applications
13 Mobile Services
3
Outline
• Motivation (Note: covered individually for each use case in the technical solution part)
• Technical solution– DIP
• Introduction and overview
• DIP Technical solution
• DIP demonstrators
– SUPER• Introduction and overview
• SUPER Technical solution
• SUPER methodology and demonstrators
• Demo/video
– seekda• Introduction and overview
• seekda! connect product
• Summary• References
3
4
DIPhttp://dip.semanticweb.org/
5
INTRODUCTION AND OVERVIEW
6
DIP – Introductory Demo/Video
(http://www.sti-innsbruck.at/results/movies/dip-promotion-movie/)
~ 9 min
7
DIP overview
Client
Services
8
DIP overview
• Let’s consider a client that want to go on holiday.• The client describes the holiday on her/his own
terms– Blue sky, white sand beach, clear water
• DIP platform acts as a broker • To fulfil user request, DIP discovers, selects,
composes and invoke services• DIP provides personalized applications on the
fly, from available services
Broker
Client
Services
• Available services: weather, hotel, travel services
9
DIP objectives
• Combine Semantic Web technology with Web Services for Semantic Web Services
• Apply Semantic Web Services as an infrastructure in real world scenarios within an organization and between organizations and its customers/partners.
• Make Semantic Web Services technology a reality.
10
DIP TECHNICAL SOLUTION
11
DIP – Overall Framework
WSMO – Web Service Modelling Ontology
WSML – Web Service Modelling Language
WSMX – Web Service Execution Environment
12
Objectives that a client wants toachieve by using Web Services
Provide the formally specified terminologyof the information used by all other components
Semantic description of Web Services: - Capability (functional)- Interfaces (usage)
Connectors between components with mediation facilities for handling heterogeneities
12
Web Service Modeling Ontology (WSMO)
13
Web Service Modeling Language (WSML)
• WSML Variants - allow users to make the trade-off between the provided expressivity and the implied complexity on a per-application basis
∩
∩
14
Web Service Execution Environment (WSMX)
• … is comprehensive software framework for runtime binding of service requesters and service providers,
• … interprets service requester’s goal to– discover matching services,– select (if desired) the service that best fits,– provide data/process mediation (if required), and– make the service invocation,
• … is reference implementation for WSMO,• … has a formal execution semantics, and• … is service oriented, event-based and has pluggable
architecture – Open source implementation available through Source Forge,– based on microkernel design using technologies such as JMX.
15
DIP Architecture
Reasoner Repository
Communication
ServiceRequester
ServiceProvider
ServiceRequester
WSML
WSML
Adapters
eB
an
king
Internet
XML
WSML
WSML
XML
ServiceProvider
Data Mediation
WSMO4J Parser Discovery
Choreography QoS Discovery
Orchestration Process Mediation
Core
WSMO Studio
16
DIP Architecture – Components (1)
• Core component– Managing exchange of messages between components
• Communication manager– Handles all external communications
• Parser– Parse WSML content of incoming messages into WSMO4j
• Discovery– Find Web services matching supplied Goals
• QoS Discovery– Find and order service on the basis of QoS parameters
• Process Mediator– Handle mismatches client and service choreographies
17
DIP Architecture – Components (2)
• Data Mediator– Handle mismatches between ontologies
• Choreography Engine– Execute behaviour described by a choreography
• Orchestration Engine– Execute the composition defined by an orchestration
• Resource Manager– Persist WSMO and operational data
• WSML Reasoner– At the heart of the architecture
18
DIP Architecture - behavioural view
19
DIP DEMONSTRATORS
20
Emergency Weather Planning
• Winter 2003 - weather chaos in southern England due to 1cm of snow.
• People spent more than 20 hours blocked on motorways
21
Emergency Weather Planning
• In an emergency situation, relevant information is needed to assist planning and decision making.
• Such information elements range from demographic data, weather forecasts and sensor data, available transportation means to the presence of helpful agents (people), etc.
• Different agencies own different relevant data and emergency related knowledge, which needs to be shared with the other partners during an emergency.
22
eMerges
• eMerges is a decision support system that assists the Emergency Office in the tasks of retrieving, processing, displaying, and interacting with relevant information, more quickly and accurately
• Using eMerges governmental agencies are able to extend their knowledge about the emergency situation they are dealing with by making use of different functionalities based on data held by other agencies which otherwise might not be accessible to them or slow to obtain.
23
eMerges Ontologies
24
eMerges Ontologies
• Archetypes ontology provides very high level abstractions (e.g. container, house, agent, etc.) to which entities from the real world have to be mapped
• HCI ontology maps an object to its particular representation. For example some interfaces need “pretty names” selecting a feature to privileged display (e.g. on hovering on the object);
25
Generic Application Structure
SWS SWS SWSSWS
SWS SWS SWS SWS
Presentation
IT systems
Organisation 1
IT systems
Organisation 2
DB DB
Web ApplicationWeb Application Web ApplicationWeb Application Web ApplicationWeb Application
Semantic Web Services (WSMX/IRS-III)
WSWS WSWSWSWS WSWS Services Abstraction
LegacySystems
SWS
26
Generic Application Structure (1)
• Legacy System layer: consists of existing data sources and IT systems provided by each of the involved governmental parties
• Service Abstraction layer: exposes the functionalities of the legacy systems as Web services, abstracting from the hardware and software platforms of the legacy systems. Whenever a new service is available at this layer, it will be semantically described and properly linked to existing semantic descriptions.
27
Generic Application Structure (2)
• Semantic Web Service layer: given a goal request this layer, will • discover a candidate set of Web services, • select the most appropriate, • mediate any mismatches at the data, ontological or
business process level, and• invoke the selected Web services whilst adhering to
any data, control flow and Web service invocation requirements
• Presentation layer: is a Web application accessible through a standard Web browser.
28
eMerges User Interface
29
eMerges Prototype Architecture
ViewEssexServices
BuddySpace Server
BuddySpace Services
Google Maps APIAJAX
AccommodationGoalEnvironment Goal
Presence Goal
ArchetypesArchetypes
SGIS-SpatialSGIS-Spatial
Em
ergency-GIS
-Dom
ainE
mergency-G
IS-D
omain
Em
ergency-GIS
-Goals
Em
ergency-GIS
-Goals
BuddySpace GoalsBuddySpace Goals
EnvironmentServices
Smart FilterServices
Google Web Toolkit
ME
T-O
ffice-Dom
ainM
ET
-Office-D
omain
ME
T-O
ffice-Goals
ME
T-O
ffice-Goals
Affordances = Goals
30
(http://www.sti-innsbruck.at/results/movies/dip-eMerges/)
~ 10 min
DIP – eMerges Demo/Video
31
SUPERhttp://www.ip-super.org/
32
INTRODUCTION AND OVERVIEW
33
SUPER – Introductory Demo/Video
(http://www.sti-innsbruck.at/results/movies/super-overview-movie/)
~ 3 min
34
SUPER
• SUPER = Semantics Utilized for Process management within and between Enterprises (SUPER)
• The major objective of SUPER was to raise Business Process Management (BPM) to the business level, where it belongs, from the IT level where it mostly resides now.
• This objective requires that BPM is accessible at the level of semantics of business experts
35
Business Process and Business Process Management
• “A business process or business method is a collection of related, structure activities or tasks that produce a specific service or product for a particular customer or customers.”
• “Business process management (BPM) is a management approach focused on aligning all aspects of an organization with the wants and needs of clients. It is a holistic approach that promotes business effectiveness and efficiency while striving for innovation, flexibility, and integration with technology”
http://en.wikipedia.org/wiki/Business_process_management
http://en.wikipedia.org/wiki/Business_process
36
Business process in a company
• Business Processes– ... drive all company‘s
activities
– ... represent the core assets of a company
– ... give decision makers control over the company’s activities
– ... deliver services faster and more efficiently to the customer
– ... allow a company to react to changing market conditions
How do I get the big
picture of my
activities?
How do I communicate my
business process in a common fashion?
How do I keep track of all evolutions in
my business?
How do I make sure my businesses get more efficient and more profitables?
37
The critical Business / IT Divide
Bridging Business-IT gap • reduce implementation costs • implementing the real requirements • faster implementation • less support requests • align implementation
Querying the Process Space • reduce costs • increase product quality • improve throughput times • less training • less support required • increase forecast accuracy
38
SUPER approach to address the critical Business / IT Divide
39
SUPER – How Semantics Help
• Semantic technology improves the utility of BPM by creating a semantic „glue“ between different layers, artefacts and models
• Links between business artefacts help to keep the „big picture“ and to improve the overall understanding of complex relationships and interdependencies
• By unifying the vocabulary and explicating differences in a structured way, semantics support the understanding of business people and technicians
40
SUPER – Scientific objectives
• Construction and assessment of technological framework for Semantic Business Process Management (SBPM)
• Acquiring new generic languages suited for representation of processes, different process models and goal description having in mind all aspects of system behaviour (e.g. costs, dependencies, constraints, other data flows, time limitations)
• Creation of automated annotation techniques of already existing BPs, their fragments, IT components, etc
• Development of process query tools• Adjustment existing reasoners to the specific needs of SUPER• Elaboration of industrial-strength mediation procedures for
automated coupling between business and IT perspectives• Augmentation of SWS foundations on the basis of new experiences
obtained from their deployment to large-scale test environments
41
SUPER – Technical objectives
• Building horizontal ontologies in aim to annotate both complete BPs and their fragments
• Assembling vertical ontologies for the chosen implementation domain
• Complete inventory of tools supporting every stage of SBPM
42
SUPER TECHNICAL SOLUTION
43
SUPER Ontology Stack
44
SUPER Ontology Stack
45
WSMO
(http://www.wsmo.org)
Objectives that a client may have when consulting a Web Service
Semantic description of WebServices: • Capability (functional)• Non-functional properties• Interfaces (usage)
Connectors between components withmediation facilities for handling
heterogeneities
Provide the formallyspecified terminology
of the information usedby all other components
Web Service Modeling Ontology
46
SUPER Ontology Stack
47
Business Domain Ontologies
• Business Functions Ontology – describes functions carried out within the company (e.g. marketing, finance, HR
• Business Process Resources Ontology – describes tangible and abstract resources required
• Business Roles Ontology – roles in the organization (e.g. Designer, Process Modeler, IT Expert, CEO)
• Business Modeling Guidelines Ontology – generic business policies and rules for domains like law, finance, etc.
48
SUPER Ontology Stack
49
• Upper-Level Process Ontology (UPO) represent high-level concepts for Business Process Modelling. It is the top-level ontology in SUPER, used as the unifying ontology for other ontologies
• Business Process Modelling Ontology (BPMO) represent high-level business process workflows. BPMO has a bridging purpose between the business level and the execution level of processes
• Semantic Event-driven Process Chains notation Ontology (sEPC) aims to support the annotation (automatic or semi-automatic) of process models created with EPC tools
SUPER Ontology Stack
50
SUPER Ontology Stack
51
• Semantic Business Process Modeling Notation Ontology (sBPMN) formalises the core subset of BPMN graphical notation
• Semantic BPEL Ontology (sBPEL) extends the BPEL ontology with a SWS based interaction model.
• Behavioral Reasoning Ontology (BRO) allows for reasoning over the behaviours of business processes using WSML axioms
• Events Ontology (EVO) is a reference model for capturing logging information utilised both by the execution engines (SBPELEE and SEE) and by the analysis tools
SUPER Ontology Stack
52
SBP ModellingTool
Semantic BPELExecution Engine
Semantic Execution Environment
SBPComposition
SBP Discovery
BusinessProcessLibrary
ExecutionHistory
SemanticWebServices
SBP Mediation
SBP Monitoring & ManagementTool
SBP AnalysisTool
Transformation
SUPER Tooling
SUPER Repositories
SUPER Execution
SUPER Platform Services
Semantic Service Bus
Deployment
Reasoner
Data Mediation
Event Sink
Protocol Binder
SUPER Architecture
53
• The central component of the architecture is the Semantic Service Bus (SSB) which provides a communication infrastructure for the SUPER components. Components communicate over the bus by sending and receiving normalized messages.
• SUPER Tooling comprises tools to support different phases of the Semantic Business Process (SBP) lifecycle:
– SBP Modeling Tool - used during the design time for SBP modeling– SBP Monitoring and Management Tool provides an up-to-date picture over the SBP and
Semantic Web services (SWS) execution state and provides simple management functionality– SBP Analysis Tool is used for Process Mining and Reverse Business Engineering (RBE)
purposes
• SUPER Repositories are used for storing artefacts which are produced, utilized and exchanged by the the SUPER components
– SBP Library stores artefacts which are created during process modelling, i.e. process models, process fragments, and process mediators
– SWS Repository stores artefacts related to Semantic Web services– Execution History stores the audit trail of the executed process instances
SUPER Architecture
54
• SUPER Platform Services comprise the basic services which provide their functionalities for all SUPER tools and components
– Transformation Services translate among different formats of SUPER artifacts– SBP Mediation resolves heterogeneity problems between different business processes– Data Mediation is responsible for handling ontology level heterogeneitie– SBP Composition combines services and processes in order to implement activities of the
process, where activities can be implemented by one or more services– SBP Discovery finds SBP candidates fulfilling criteria specified as WSMO Goals– SBP Reasoner provides process behavioural logic-based inference engine capable of
reasoning with SUPER ontologies
• SUPER Execution comprise two execution environments/engines:– Semantic BPEL Execution Engine is a BPEL 2.0 compliant process engine, which
supports the extensions of BPEL4SWS and is integrated into the Semantic Service Bus (SSB)
– Semantic Execution Environment (SEE) enables discovery, selection, mediation, invocation and interoperability between Semantic Web services (SWS). SEE is a middleware operating on WSMO descriptions enabling flexible interaction between Service Requesters and Service Providers
SUPER Architecture
55
SUPER METHODOLOGY AND DEMONSTRATORS
56
SUPER Methodology Framework
Text Text
Text
Semantic Business Process
Analysis
Semantic Business Process
Execution
Semantic Business Process
Configuration
Strategic Semantic Business Process Management
Ontological Foundation
Semantic Business Process
Modelling
The SUPER methodology is a set of phases, methods and techniques to perform activities using SUPER technologies. Like a traditional BPM methodology, the SUPER methodology owns a proper business process “life cycle”, that is enriched with the semantic connotation of the overall SUPER framework.
57
Semantic Business Process Modelling
• Semantic Business Process Modelling (SBPM) is the first step of the SUPER Life Cycle
• SBPM is concerns with a streamlined, comprehensive, and easy- to-use representative model of the real enterprise business processes
• Development of the Business Processes
Model based on the Business Process
Modelling Ontology (BPMO)
• Use of a Semantic Process Modelling Environment– WSMO Studio– Integrated BPMO Editor
58
Example: TID Prototype
59
TID Modelling – Demo/Video
(http://www.sti-innsbruck.at/results/movies/tid-modeling-tool-developed-in-super/)~ 5 min
60
Benefits of SUPER Modelling
• Business Process Modelling Notation (BPMN) independence (BPMO representation)
• Discovery of existing Business Processes exploiting the semantic information– Search on specified Business Function, Domain and Patterns– Search on specified Business Goals, KPIs and Business Rules
• Automatic validation and simulation of the BPM• Better readibility of models through a clear semantic
61
BPMO Editor Demo/Video
(http://www.sti-innsbruck.at/results/movies/bpmo-editor-part-of-wsmostudio-developed-in-super/)
~ 10 min
62
Semantic Business Process Configuration
• Semantic Business Process Configuration (SBPC) is the second phase of the SUPER SBP Life Cycle. It uses the outputs of the SBP Modeling phase and provides inputs for the third phase, the Semantic Business Process Execution
• During this phase Modelled Business
Processes are configured
63
Semantic Business Process Configuration
Semantic Business Process Configuration steps:
1. Derive sBPEL from BPMO
This step enables the translation from the BPMO instance (coming from the SBP Modeling phase) to an sBPEL ontology
2. Search for possible SWS
This step consist in discovery of SWS. Even if the services will be executed in the SBP Execution phase, an early service discovery could be extremely useful to reduce the effort of the service selection before the execution
64
Semantic Business Process Configuration
3. Examine potential data mismatches
In this step data have to be examined to identify potential data mismatches.
4. Define data mappings and mediatiors
If potential data mismatches are identified in the previous step interface mappings and data mediators have to be created
5. Validate and refine the process
In this step the process is validate and potentially refined. The validation is seen as a sort of “compiler” that checks the correctness of the semantic process description before the execution of the process.
65
Semantic Business Process Execution
• Modeled and configured Semantic Business Processes are executed
• Execution history for SBP Analysis is produced
• Automates business activities• Minimizes time-to-offer• Supports
– Execution of semantic BPEL processes (BPEL4SWS)
– Discovery and execution of Semantic Web Services (SWS)
Text
Semantic Business Process
Execution
Semantic Business Process
Modelling
Semantic Business Process
Configuration
Semantic Business Process
Analysis
66
Semantic Business Process Execution Scenario
After the process execution has been finished, the result is returned to the user. After the process execution has been finished, the result is returned to the user.66
SUPER RepositoriesSUPER Execution Environment
SemanticWeb
Services
Achieve Goal
Discover Service
6
2
4
35
1
Return result to engine
Semantic BPEL
Execution Engine
(SBPELEE)
Semantic Execution
Environment(SEE)
Semantic Web Service(SWS)
SUPER Tooling
MonitoringTool
ExecutionHistory
A user initiates the semantic BPEL process by sending a service request through the Semantic Service Bus to SBPELEE.
A user initiates the semantic BPEL process by sending a service request through the Semantic Service Bus to SBPELEE.
11
Invoke Service
Return ResultRequest Service
Semantic Service Bus(SSB)
SBPELEE delegates the invocation of SWS to SEE by passing the WSMO Goal to it.
SBPELEE delegates the invocation of SWS to SEE by passing the WSMO Goal to it.
22 SEE queries the SWS repository to discover the desired SWS. SEE queries the SWS repository to discover the desired SWS.33 SEE invokes the selected SWS. SEE invokes the selected SWS.44 SEE returns the result of “Achieve Goal” to SBPELEE. SEE returns the result of “Achieve Goal” to SBPELEE.55During the execution, execution events are published to Execution History for persistence and to the Monitoring Tool for tracking process executions.
During the execution, execution events are published to Execution History for persistence and to the Monitoring Tool for tracking process executions.
67
Semantic Business Process Execution Scenario
• Step 1: A user initiates the semantic BPEL process by sending a service request through the Semantic Service Bus to Semantic BEPEL Execution Engine (SBPELEE ).
• Step 2: SBPELEE delegates the invocation of SWS to Semantic Execution Environment (SEE) by passing the WSMO Goal to it.
• Step 3: SEE queries the SWS repository to discover the desired SWS.• Step 4: SEE invokes the selected SWS.• Step 5: SEE returns the result of “Achieve Goal” to SBPELEE.• Step 6: After the process execution has been finished, the result is
returned to the user.
• During the execution, execution events are published to Execution History for persistence and to the Monitoring Tool for tracking process executions.
68
Cu
stom
er
Ne
xco
m S
yste
m
Receive Service Request
Gro
up
Cu
stom
ers
Sa
les
Bill
ing
Receive Price and Quality
Receive Offer
Su
pp
lier
Su
pp
liers
No
Yes
Receive Request for Price and
Quality
Obtain Price and Quality data
Check for Price and Quality Send Offer
Send Service Request
Receive Price and Quality
Send Preference
Receive Price and QualityPreference
Send Price and Quality
Negotiate Price and Quality
Supplier Match?
Example: Nexcom Customer Order Management Process
SupplierSupplierSupplier
1 6
2 5
1 6
Supplier exposes its process as SWS.Supplier exposes its process as SWS.
Nexcom process is deployed as a semantic BPEL processNexcom process is deployed as a semantic BPEL process
Customer uses a client application to start the Nexcom processCustomer uses a client application to start the Nexcom process
69
Benefits from SUPER SBP Execution
• Nexcom Use case requirements addressed by the SUPER SBP Execution phase– Supplier matching supported by Semantic Web Service
discovery and invocation from within semantic business processes
– Allows for more flexible traffic routing– Automates supplier matching and traffic routing process taking
into account all existing suppliers– Minimizes time-to-offer
70
SBP Execution Demo/Video
(http://www.sti-innsbruck.at/results/movies/sbp-execution-developed-in-super/)
~ 11 min
71
Semantic Business Process Analysis
• Analysis of executed processes• Support of various analysis goals
– Overview over process usage– Detect business exceptions– Detect technical exceptions– Compare As-Is with To-Be
• Analysis methods– Semantic Process Mining– Semantic Reverse Business Engineering
Semantic Business Process
AnalysisSemantic
Business Process Modelling
Semantic Business Process
Execution
Semantic Business Process
Configuration
72
Semantic Reverse Business Engineering (RBE)
• Scenario based analysis with predefined content to ensure continuous business improvement
– As-Is-AnalysisProvide details and statistics about executed processes
– Exception analysisFocus on business exceptions (deviation from the standard processes)
– Standardisation & HarmonisationCheck compliance of processes between organisational units or with predefined guidelines
– User & Role analysisCheck user and role behaviour and authorizations
How do I get the relevant information
to redesign and improve my business
processes?
7373
Business Question Repository
Business Function OntologyRBE Ontology
Sales Process
ExceptionAnalysis
Scenario Based Analysis
How many sales orders were cancelled?
Which sales orders are locked for further processing?
How many sales orders are delayed?
I am interested in all exceptions of the sales process
ExecutionHistory
Repository
Analysis Results
Analysis Results
Where are the bottlenecks in the sales process?
Process MiningProcess Mining
Only business questions semanticallyassigned to Exception Analysis and to the Sales Process are to be selected
Only business questions semanticallyassigned to Exception Analysis and to the Sales Process are to be selected
Business questions are executed on the Execution History Repository(log file) either directly or through Process Mining
Business questions are executed on the Execution History Repository(log file) either directly or through Process Mining
The query results are formatted and aggregated for the business user
The query results are formatted and aggregated for the business user
74
• The business user who wants to perform a specific analysis needs to select the relevant business questions (BQ). If we had just a few questions, this operation could be performed manually. But since we deal with a rich set of BQs, we need a smarter way to select them.
• Therefore the business user has just to select the relevant concepts within the SUPER ontologies (e.g., he wants to perform an exceptional analysis, within the sales processes)
• In this way, the analysis tool is able to automatically select the BQs annotated with those concepts.
• These Business Questions are either directly executed on the execution history repository or “brought” to the process mining environment.This is dependent on the question. Some business questions can directly be answered (RBE approach), some business questions can only be answered using process mining.
• Once the respective queries (either triggered by the Business Question repository directly or by the Process Mining environment) are executed on the Execution history repository the query results are formatted and aggregated for the business user.
Scenario Based Analysis
75
Analysis Results
Successful SalesOrders
Cancelled SalesOrders
► Get overview about system usage► Find out exceptions within process flow► Check conformance to defined Process model► Find bottlenecks► Get basis information to apply 6-sigma methodology
How many sales orders were cancelled?
Which sales orders are locked for further processing?
76
SBP Analysis Demo/Video
(http://www.sti-innsbruck.at/results/movies/sbp-analysis-developed-in-super/ )
~ 11 min
77
DEMO/VIDEO
78
SUPER Demo/Video
http://www.sti-innsbruck.at/results/movies/super-integrated/~ 20 min
7979
80
INTRODUCTION AND OVERVIEW
81
seekda!
• Mission: development and marketing of trendsetting, internet-based web services
• Core Competence: Commercialization of Web Services
• Business activities: automated locating, creating and structuring of new business models, preparing for eCommerce, as well as the integration of existing or the bundling into new, more comprehensive services (e.g.: unified-messaging-solutions).
• Premium, customized business solutions:
– Online distribution multi channel solution (seekda! connect)– Conference and congress management tool (seekda! conferences)– The first web service search engine worldwide
82
seekda!
seekda! uses technologies and provides innovation in the following areas:1. Service Oriented Architectures2. Web of Services
3. Web Services Search Engine
4. Semantics and Service Commerce
83
1. Service Oriented Architectures
• IT solutions shift from monolithic
systems towards Service Oriented
Architectures• Organizations require on-demand
external services• Interface vs. Implementation
– I only do care about interface (description) but do not care about implementation (program) as somebody does it for me
• However, current services market:– still not transparent– provider and/or technology centric– mainly atomic services and not bundles/solutions– low technical quality of interfaces, high setup and migration costs
84
1. Service Oriented Architectures
• Current change: Web of pages → Web of services– there is already a considerable number
of publicly available services– …but users need to be aware of the
existence and the features of a service • UDDI standard did not prevail• Specific portals: access to restricted sets of
registered services
85
2. Web Services Search Engine (1)
• Search Engine for Web Services– fully automated focused crawling process
– aggregating information from multiple sources into a semantic model
– efficient means for finding services
– community features enabling understanding and selecting right services
86
2. Web Service Search Engine
87
2. Web Service Search Engine
Focused Crawling
• Core Issues– Good seed URLs– Assign score for resource content– Guess content based on URI pattern
• Things to look after:– IP politeness– spam/crawler traps– bandwidth, storage, cpu, ...
• Divide and Conquer– Partition URLs to multiple machines– Separate Frontier to multiple queues
88
2. Web Service Search Engine
Crawling Architecture
• Collect meta data automatically from various source• Create Object based search for services
89
2. Web Service Search Engine
Analyzing Data
• http://seekda.com/providers/amazon.com/S3
• Amazon S3 provides a simple web services interface that can be used to store and retrieve any amount of data, http://www.amazon.com/gp/browse.html?node=16427261
• Commercial Service, Computing
• Pricing$0.15 per GB-Month of storage used http://calculator.s3.amazonaws.com/calc5.html
• Terms of Servicehttp://www.amazon.com/AWS-License-home-page-Money/b/ref=sc_fe_c_0_16427261_10/104-7739021-5554311?ie=UTF8&node=3440661&no=16427261&me=A36L942TSJ2AJA
• Related Services
– boxNetStorage (box.net)– OnlineStorage (digitalbucket.com)
90
2. Web Service Search Engine
Web Services Domains
91
3. Service Commerce
• How to enable Service Commerce – towards a Web Service Marketplace– one-stop-shopping across multiple service providers
– aggregation & mediation - create service bundles according to users needs
– provide applications that utilize services (or service bundles) e.g. on-demand printing services, conference organization, etc.
– combine services while checking technical and contractual compatibilities
– technical integration simplified by using semantic technologies
92
The Semantic Magic – What is needed?
• Discovery– Automated focused crawling – Aggregating information from multiple sources into semantic model– High quality without relying on manually maintained registry
• Mediation– Service interfaces are mapped to ontologies– Mapping rules specified on a semantic level– Enables runtime exchange of similar services
• Bundling/Composition– Semantic descriptions enable semi-automatic creation– Technical service integration done by marketplace– Consuming a bundle as easy as an atomic service
93
Mediation
Design Time
Mappings to target vocabulary (Pegasus)
Semi-automatedmappings generation
Mediation Engine
Repositories and Data Stores
MessageMappings
Semantic enriched vocabulary (OTA)
Run time
Mappings to target vocabulary (Tiscover)
Mappings to target vocabulary (Expedia)
Message instance (Expedia)
Message instance (Tiscover)
Target Message instance (Pegasus)
Source vocabulary(OTA)
Domain Expert
Mappings to target vocabulary (n)
Mappings to target vocabulary (PMS)
lookup
Source message instance (PMS)
Map Central Format(OTA)
Mappings to target
vocabulary (PMS)
MapMappings to target
vocabulary (Expedia)
Mappings to target
vocabulary (Tiscover)
Mappings to target
vocabulary (Pegasus)
lookup
94
Composition – the Process
95
Mediation & Bundling Example
• There are many equivalent services capable to fulfil a goal of a customer (e.g. sending SMS)
• Every provider uses its own format to represent SMS service – mediation is then required
• Benefits for customer: always the best (the most suitable) service is selected for invocation
96
3. Service Commerce
SaaS
• Software as a Service (SaaS)– "Software deployed as a hosted service and accessed over the Internet”
(Microsoft)
– SaaS applications are typically contrasted with on-premise applications
– Network tends to mean the Internet, leveraging Web technologies tends to be essential
Goldman Sachs, November 2007
• 23% projected annual growth• $21.4 billion industry by 2011
97
3. Service Commerce
SaaS - Business Case for Providers
• Reduce the substantial costs of code delivery to the customer• Expand the potential customer base• Revamp traditional business models to a more Internet focused
approach• Offer better online services and information to customers• Limit the costs of configuring software for customers, suppliers and
internal users
98
3. Service Commerce
SaaS - Business Case for Customers
• Why buy when you can rent?• Transforming IT departments from application developers to application
users• Greater flexibility and scalability• An expectation for unleashing new value of previously isolated data
silos and functionality
99
seekda! connect product
100
Why seekda! connect
• Hotels use various distribution channels.
• Daily maintenance of right balance of rooms availability across multiple channels does not scale.
• Average time for hoteliers required to maintain a profile of a medium size hotel at one portal takes between 5 to 15 minutes a day.
• An effort of maintaining hotel’s profile on 10 portals would require then at least 2 hours of work.
101
What (1)
• The multi-channel-solution for hotel-industry
internet distribution
seekda! connect
seekda! IBE
102
What (2)
• seekda! connect open new dimensions of online distributions for hoteliers allowing them to manage various distribution channels with only one tool.
103
What (3)
• Additional services (car rentals, insurance,
airport shuttles) offered directly at hotel’s Website
104
What (4)
• additional services (concert tickets, ski passes,
sport offers) offered through mobile devices to
travelers, not only before, but also during the trip.
105
How (1)
• seekda! connect Service Platform provides
advanced means for authentication, authorization, accounting, mediation, process definition, process execution and monitoring.
106
How (2)
• Engineers and domain experts from the company
define mappings allowing to generate messages suitable
to be delivered to destination systems• This work is heavily based on OTA, which is industry standard for tourist
industry
107
How (3)
• seekda! connect open API defines the proper
semantics to simplify data, process and protocol mediation.
Person
Tourist Hotelier
Hotel
isA – hierarchy (taxonomy)
name email
address
amenitiesHotel location
visits manages
108
Competition (e-tourism market)
Innovative
Local (Europe)
Conventional
IndividualProperties
Global (US) Chain Hotels
109
Competition (marketplaces)
• StrikeIron – the biggest marketplace for commercial Web services; – seekda’s Search Engine make the market more transparent then StrikeIron– StrikeIron does not allow to share subscription between services
• widgetbox – Funding: $1.5 million from Hummer Winblad; Advertises itself as an "open web widget marketplace and syndication platform”:
– their widgets are all still free, but the intention is to provide support for commercializing widgets.
– targets in particular bloggers and website owners to deploy the widgets– has pre-build integration with some blog engines and content management systems
to lower the barrier for people to deploy their widgets.
• SpringWidget - internally funded; it is also a widget marketplace, – puts a stronger emphasis on having widgets available on many platforms (blogs,
website, desktop) with the cost of needing a stronger integration of the widget with the SpringWidget's platform.
110
Competition (search engines for Web Services)
• programmableweb about 400 APIs, seems non-commercial, but good content and can be used for many ideas...
• StrikeIron global directory• Xmethods – about 400 services, simple interface• Merobas• webrpc - very small set of services, however more community features
and not only WSDL• SoaHub – portal about WS• WSindex (small set of WS)• wbslogger (set of WS in different categories and languages, user can
rate and comment entries)
111
SUMMARY
111
112
DIP achievements
• DIP provides an Open Source Architecture for Semantic Web Services– DIP Architecture, DIP API, WSMO4J
• DIP provides a set of comprehensive tools– WSMX, IRS-III, WSMO Studio, Hybrid Reasoning tool
• Real Use Case Implementations have been developed in DIP– Diverse scenarios e.g. eMerges for emergency weather planning
• Standards Impact– W3C Member Submissions, OASIS
113
SUPER achievements
• SUPER bridges the gap between Business experts and IT experts in setting up new products and processes.
• SUPER provides a new set of integrated BPM tools for – Modeling– Automated Composition of Processes
• SUPER uses Semantics to gain a new level of automation for the modeling and configuration of business processes.
• SUPER tools are based on open standards to guarantee independence from particular vendors.
• SUPER enables lower development costs and short time-to-market for new services and products.
114
seekda
• Mission: development and marketing of trendsetting, internet-based web services
• Core Competence: Commercialization of Web Services
• Business activities: automated locating, creating and structuring of new business models, preparing for eCommerce, as well as the integration of existing or the bundling into new, more comprehensive services (e.g.: unified-messaging-solutions).
• Premium, customized business solutions:
– Online distribution multi channel solution (seekda! connect)– Conference and congress management tool (seekda! conferences)– The first web service search engine worldwide
115
REFERENCES
115
116
References
• Mandatory reading:• http://dip.semanticweb.org/
• http://www.ip-super.org/
• http://seekda.com/
• Further reading:– DIP• Deliverables: http://www.ip-super.org/content/view/32/66/
• Ontologies: http://www.ip-super.org/content/view/129/136/
• Tools: http://www.ip-super.org/content/view/196/163/
• Demonstrators: http://www.ip-super.org/content/view/192/144/
117
References
– SUPER:• Deliverables: http://www.ip-super.org/content/view/32/66/
• Ontologies: http://www.ip-super.org/content/view/129/136/
• Tools: http://www.ip-super.org/content/view/196/163/
• Demonstrators: http://www.ip-super.org/content/view/192/144/
– seekda:• http://connect.seekda.com/en
• http://conferences.seekda.com/en/
• http://webservices.seekda.com/
118
References
• Wikipedia and other links:– http://en.wikipedia.org/wiki/Business_Process_Modeling_Notation– http://en.wikipedia.org/wiki/Semantic_Web_Services– http://en.wikipedia.org/wiki/Business_process– http://en.wikipedia.org/wiki/Business_process_management– http://en.wikipedia.org/wiki/WSMO
119
Next Lecture
# Title
1 Introduction
2 Web Science
3 Service Science
4 Web services
5 Web2.0 services
6 Semantic Web
7 Web Service Modeling Ontology (WSMO)
8 Web Service Modeling Language (WSML)
9 Web Service Execution Environment (WSMX)
10 OWL-S and others
11 Light-weight Annotations
12 Applications
13 Mobile Services
120
Questions?