Administration Arranged Registering Administration Disclosure and Creation.

Uploaded on:
AI Planning for Service Composition. General constraints for administration structure ... Total for successive structure, Max for parallel arrangement, Min for decisions ...
Slide 1

Administration Oriented Computing Service Discovery and Composition

Slide 2

Service Discovery UDDI Discovery taking into account WSDL data WS-Discovery Provide an interface for administration revelation Define a multicast disclosure convention Limitations: No administration liveness data, restricted administration portrayal Universal Plug and Play (UPnP) Enables dynamic systems administration of astute machines, remote gadgets, and PCs Not precisely web administrations, yet is another type of institutionalization for physical frameworks

Slide 3

WS-Discovery Extend UDDI to make it conveyed WS-Discovery multicast message sorts: Hello: Sent by a Target Service when it joins a system Bye: Sent by a Target Service when it leaves a system Probe: Sent by a Client hunting down a Target Service Search by Type and/or Scope Resolve: Sent by a Client hunting down a Target Service by name Already know the objective administration by name, yet may not know the correspondence subtle elements Response uni-cast message sorts: Probe Match: a Target Service coordinates a "Test" Resolve Match: a Target Service coordinates a "Purpose"

Slide 4

UPnP Devices When entering a system, register to the control moment that leaving the system, data the control point Control moment that entering the system, look for gadgets

Slide 5

UPnP Device determination Device name, seller name, URL for the gadget, and so forth. Summons, activities, parameters, and so on. Current state data Interactions Control: Control point sends charges/activities (in XML) to initiate the gadget Event: Control point can ask for gadget to send redesigns when a few variables (indicated in the occasion) are overhauled Device can acknowledge the solicitation and react with an occasion term Presentation If the gadget has a URL, control point can demand and get it Some gadgets can be controlled through the URL interface

Slide 6

UPnP Compare to UDDI Similar, yet has an extra communication highlight Lack of semantics, difficult to create the gadgets together to accomplish a customer objective Can we wrap gadgets into abnormal state administrations and utilize the OWL-S advancements to add semantics to gadgets?

Slide 7

Semantic Web Service Discovery Semantic pursuit format User\'s administration prerequisites Semantic web administration portrayals A similitude based coordinating plan (only an illustration) The Search Template is coordinated against an arrangement of applicant Web administrations in a registry and the match Scores are figured Overall comparability = Weighted normal of syntactic closeness and Functional likeness (standardized whole of operational closeness) Syntactic likenesses = Weighted normal of the Name comparability and Description closeness Operation similitude = Weighted normal of syntactic likeness, theoretical similitude, and I/O closeness … (further decay the closeness definitions)

Slide 8

Semantic Web Service Discovery Example seek layout

Slide 9

Semantic Web Service Discovery Example competitor administration

Slide 10

Semantic Web Service Discovery Similarity definitions Best match! Contrast a craved administration name and the names of a few applicant benefits Best match! Contrast a coveted operation and every one of the operations of a hopeful administration

Slide 11

Service Composition How to assemble administrations to accomplish the wanted objective Same old issue Design designs has appeared to be viable in programming plan Many modern endeavors on SOA outline designs What is an example "A answer for an issue in a context"? Every example depicts an issue which happens again and again ... and afterward depicts the center of the answer for that issue, in a manner that you can utilize this arrangement again and again Research Pattern based examination considers how to indicate designs, i.e., how to determine the issue, arrangement, impacts Semantic Web administration creation group consider AI arranging systems for piece thinking

Slide 12

AI Planning for Service Composition Planning move(x,y) Pre-condition: clear(x) and clear(y) Effect: on(x,y) and clear(x) Delete impacts: on(x,?), clear(y) Always clear(table) Does not struggle with on(x, table) Initial state on(c,table) on(b,table) on(a,b) clear(a), clear(c) Goal state on(a,table) on(c,a) on(b,c) clear(b) Composition move(a,table) move(c,a) move(b,c) B A C B C A

Slide 13

on(c,table) on(b,table) on(a,b) clear(a) clear(c) clear(table) on(a,c) move(a,c) move(a,table) on(c,table) on(b,table) on(a,b) clear(a) clear(c) clear(table) on(a,table) . . . move(c,a) AI Planning for Service Composition on(c,table) on(b,table) on(a,b) clear(a) clear(c) clear(table) . . . Arranging is not the same as other hunt calculations - which by and large taking a quantitative measure Planner includes state calculation and upkeep

Slide 14

AI Planning for Service Composition Map semantic web to the arranging space Definitions for web administrations Syntactical definition: I/O parameters Semantic definition: pre-condition and impacts Supported in OWL-S and WSMO Map administrations to activities in arranging area Pre-condition/impacts of the administrations turn into the pre-condition/impacts of the activities I/O definitions are deciphered t o the pre-condition/impacts Map the issue to the arranging space Define the objective for the issue Define the underlying certainties

Slide 15

AI Planning for Service Composition Limitations of conventional organizers for administration piece Atomic activities with deterministic impacts, just ready to produce successive arrangements Conditional, iterative arranging: C onstruct an arrangement with branches, considering all conceivable nondeterministic impacts and possibilities into record Complete learning of the world, full perceptibility Conformant arranging: F ind an arrangement which works in any underlying circumstance or inadequate information Contingency arranging: Consider all conceivable nondeterministic impacts or replan when surprising circumstance happens

Slide 16

AI Planning for Service Composition General restrictions for administration organization Pre-conditions and impacts, introductory states, and objectives are for the most part basic conjunctions of suggestions Can true web administrations be effortlessly indicated in view of these? Has been a center issue in SE for a long time!!! Will it work now? Scaling issues There might be a great many administrations each with various ports Even more awful in digital physical frameworks A ton gadgets with comparable functionalities Solutions Hierarchical pursuit: First utilize catchphrase based inquiry to sift through far-fetched activities, then utilize organizer to investigate the conceivable activities Service metaphysics: Categorize benefits and determine administration relations utilizing a philosophy

Slide 17

AI Planning for Service Composition General restrictions for administration structure Assume a static and limited arrangement of activities In SE, an issue can be decayed and after that locate the relating segments Research work attempting to handle halfway arranging with missing activities Interaction with clients Planer better be more blended activity learning building issues Efficiently and viably connecting with XML based data This ought to be the least complex issue among the numerous issues

Slide 18

QoS in Service Composition QoS (nature of administration) Nonfunctional properties to be fulfilled E.g., accessibility, execution, value, unwavering quality Service sythesis with QoS contemplations Find the best support of meet client QoS prerequisites Need to indicate customer QoS necessities First, need to characterize what QoS is (the properties of concern) Need to know the QoS properties of the administrations For a solitary administration, QoS properties can be measured For a composite administration, how to infer the properties of the created administration? For a few properties, property conglomeration can be extremely troublesome Also, need to comprehend the cooperation practices among administrations Decision making: which administrations to choose?

Slide 19

QoS in Service Composition Need to have A formal procedure to do this methodicallly, from QoS detail to arrangement, to finish the choice An understanding between the included elements to guarantee that the arranged QoS terms are exercised  Service Level Agreement (SLA) WS-Agreement Provides the particular gauges for SLA between the customer and the administration suppliers Dynamically settled and progressively oversaw QoS Use QoS philosophy for QoS determinations, transaction, and administration

Slide 20

Factory Negotiation make() Ops: terminate(limits) negotiate(...) ... SDEs: arrange() Terms Status Related Agrmts. Moderator WS-Agreement Negotiation Layer Agreement Layer Service Layer Factory: makes the occurrence Factory Agreement make() Ops: terminate(limits) inspect(query) ... SDEs: review() Terms Status Related Agrmts. Chief Factory make() Policy Application Instance foo() Consumer Provider

Slide 21

WS-Agreement Negotiation layer Provides a Web administration based bland arrangement capacity Newly included (unique just has two layers) Negotiation state move:

Slide 22

Name Context WS-Agreement Context Agreement initiator, responder, expire time, and so on. Administration Terms Identify the particular administrations to be given Guarantee Terms The administration levels that the gatherings are concurring upon Can be utilized for observing Agreement Terms Service Terms Guarantee Terms

Slide 23

WS-Agreement hasGuaranteeTerm GuaranteeTerm hasBusinessValue A surety term has a degree – e.g. operation of administration hasScope hasObjective hasCondition Scope BusinessValue ServiceLevelObjectives Qualifying Condition hasReward An assurance term may have gathering of administration level goals e.g. responseTime < 2 seconds There may be business values connected with every certification terms. Business values incorporate significance, certainty, punishment, and prize. e.g. Punishment 5 USD Reward An insurance term may have a qualifying condition for SLO\'s to hold. e.g. numRequests < 100 Predicate hasPenalty hasImportance Penalty Parameter Unit Importance Value ValueExpression Unit Value ValueUnit ValueExpression OWL philosophy Asse

View more...