1 / 29

Behaviour, Time and Viewpoint Consistency: Three Challenges for MDE

Behaviour, Time and Viewpoint Consistency: Three Challenges for MDE. José E. Rivera, José R. Romero, A. Vallecillo Universidad de Málaga. Complexity (i). [Borrowed from Dov Dori’s Tutorial on SysML Modeling at TOOLS 2008]. Maininframe. PC/NT apps. Unix apps. Vendor Setup.

Télécharger la présentation

Behaviour, Time and Viewpoint Consistency: Three Challenges for MDE

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Behaviour, Time and Viewpoint Consistency: Three Challenges for MDE José E. Rivera, José R. Romero, A. Vallecillo Universidad de Málaga ChaMDE@MoDELS2008

  2. Complexity (i) [Borrowed from Dov Dori’s Tutorial on SysML Modeling at TOOLS 2008] ChaMDE@MoDELS2008

  3. Maininframe PC/NT apps Unix apps Vendor Setup 3rd Party Interface Vendor Process Servers Maintenance Budget (Imaging) Analysis Tool Printer S20-Sales Maintenance Polling Printer PO Insertions S01 - Sales AIS Reports Orders Corrections AIS Calendar I06 Warehouse Due Dates Print Costing Management General Invoice App Depository Broadcast Maintenance Stores & Mrkts Banks Filter Smart Plus I17 Customer Perceived M03 - Millennium 3.0 Smart Plus UAR - Universal Account In-Stock NEW Soundscan Sterling VAN Launcher Reconciliation NPD Group Roadshow Mailbox (Value) Mesa Data AIG Warranty Guard I03 Return to I13- Auto I15 Hand Scan Vendor Replenishment M02 - Millennium S06 - Credit App I06 - Customer Apps Order P15 EES Employee Change Notice L02-Resource E13 Scheduling Fringe PO E3 Interface E01-EDI P14 On-line New Hire Entry S04 - Sales Posting S07 - Cell Phones P16 - Tally Sheet D01 Post Load Billing P09 - P17 Cobra Cyborg Equifax CTS Stock Options S09 - Digital I12 Entertainment Satellite Software Prodigy System A04 - Cust L01-Promo Refund Chks 1 Analysis Banks - ACH and Pos to Pay AAS Plan Administrators V02-Price P01- Resumix B01 - Stock Spec Source (401K, PCS, Life, Marketing Washington, Employee Status SKU Tracking Support Unicare, Solomon RGIS, Masterfile Smith Barney) Ntl Bus Systems Frick S11 - ISP CTO2 Co Tracking I10 Cycle Physical I04 Home Inventory ACH Deliveries I01 PO V04-Sign Receiving System Scorecard - HR U18 - CTO I02 - POS X92-X96 Transfers Host to AS400 S03-Polling Communication I11 Price I09 Cycle Counts Testing Supplier S08 - Vertex Compliance NPD, S02 - Intercept Sales E02-Employee Layaways SoundScan Tax Purchase Spec Source SKU Performance V03- Mkt Reactions L60 MDF P09 Coop G02 - General I05 V01-Price Management SKU Selection Bonus/HR Ledger Inventory Info System Tool I35 - CEI K02 Customer Repair ASIS Arthur Planning I35 Early Warning Tracking System I18 Rebate SKU Rep Transfer I55 SKU I07 Purchase Store Information Order Ad Expense ELT Monitor PowerSuite Store Scorecard Journal Entry Tool Kit Sign Texlon 3.5 A05 - AP System NARM I14 Count Corrections Store Budget Reporting INVENTORY CONTROL APPS - PC INVENTORY CONTROL APPS - PC ACCTS REC APPS - PC Valley Media Code Alarm DPI/CPI 990COR B02 Merchandise OTHER APPS - PC Debit Receivings IC Batching U16-Texlon Bad Debt CopyWriter's Analysis AP - Collections/Credit Devo Sales Inventory Adj/Count Correct BMP - Bus Beneficial Fees Workspace TM - Credit Card DB Display Inventory Inventory Control Reports Beneficial Reconcile performance Mngt In Home Inventory Levels JEAXF EDI Junkouts Inventory Roll JEBFA Coordinator Merchandise Withdrawal Merchandise Withdrawal JEBKA PSP Promo Credits Open Receivings JEDVA Merch Mngr Approval RTV Accrual PI Count Results JESOA Batch Forecasting Shrink PI Time Results from Inv AIMS JEVSA Connect 3 AP Research - Inv Cntrl Price Protection JEVSF Ad Measurement AP Research-Addl Rpts Sales Flash Reporting NSF AIMS Admin In-Home Book to Perpetual Inventory Shrink Reporting TeleCredit Fees Data Warehouse Close Out Reporting Repair SKU Gross Margin Connect 3 Connect 3 Cellular Computer Intelligence Data (Interfaces to and from the SKU Shrink Level Detail Reports PDF Transfer Rollover Count Corrections USM Data Warehouse are not AIMS Cross Ref for VCB Dnlds VCB Downloads displayed on this diagram) Ad S05 - House Reporting Warranty Damage Write Off Launcher Charges Billing Debit Receivings DFI Vendor Database System Display Inventory Reconcile US Bank Recon Optika Display Inventory Reporting File C02 - Capital Projects ICMS Credit SiteSeer Cash Receipts/Credit F06 - Fixed Misc Accounting/Finance Apps - PC/NT Assets COBA (Corp office Budget Assistant) Star Repair Cash Over/ PCBS(Profit Center Budget System) Short Merchandising Budget Complexity (ii) Design of a real Retail application ChaMDE@MoDELS2008

  4. MDE is more than Conceptual Modeling! • Current DSLs • Toy-ish • Unanimated (mostly static) • Limited analysis capabilities • Almost inexistent tool support for • Simulation • Analysis • Estimation • Quality evaluation and control • ... • Almost inexistent proven methodologies • For neither development nor modernization ChaMDE@MoDELS2008

  5. We need to be able (at least) to: • Deal with both the accidental and the essential complexity of complex systems • Use separate viewpoints to specify systems (each viewpoint uses its corresponding DSL) • Check the consistency of multi-viewpont specifications • Animate models • Explicitly define behavioral semantics of DSLs so that models can be understood, manipulated and maintained by both users and machines • Define different semantics (separate concerns) • Analyze models • Add Non-Functional Properties to DSLs • Connect DSLs to Analysis tools ChaMDE@MoDELS2008

  6. Use of models to connect the tools [Borrowed from Russell Peak presentation at OMG, 2007] ChaMDE@MoDELS2008

  7. Challenge 1. How do we add behavioral descriptions? • …to animate models (i.e., execute them) • …to be able to conduct simulations • …to be able to perform different kinds of (automated) analysis ChaMDE@MoDELS2008

  8. Anatomy of a DSL ChaMDE@MoDELS2008

  9. Abstract and concrete syntax ChaMDE@MoDELS2008

  10. Anatomy of a DSL (ii) ChaMDE@MoDELS2008

  11. Bridges between Semantic Domains ChaMDE@MoDELS2008

  12. Bridges between Semantic Domains • Precise semantics • A set of (equivalent) notations • A set of Analysis Tools • Underlying logic ChaMDE@MoDELS2008

  13. How to implement the Mappings? • As Model Transformations!!! • Types • Domestic • Horizontal • Vertical • Abstracting • Refining • Pruning • Forgetful • … ChaMDE@MoDELS2008

  14. ChaMDE@MoDELS2008

  15. One way of adding behavioral semantics • Using in-place model transformations l:[NAC] x LHS → RHS ChaMDE@MoDELS2008

  16. Q4. How do we analyse models? • Crossing the bridges!!! ChaMDE@MoDELS2008

  17. Challenge 2. Specifying more NFP to analyze models • First one is Time… • But there are others… • Probabilities • Resource consumption • SLAs • … • How to add them to our behavioral specifications? • How to connect them to existing analysis tools? ChaMDE@MoDELS2008

  18. One way to add Time • Using in-place model transformations (same) • But adding the duration of the action t l:[NAC] x LHS → RHS ChaMDE@MoDELS2008

  19. Precise Semantics of Timed Rules • Defined by a Semantic Mapping to Real-Time Maude • This makes models amenable to formal analysis using the Real-Time toolkit! ChaMDE@MoDELS2008

  20. Model-driven Run-time monitoring [MDD-MERTS Spanish project TIN2008-03107, 2009-2011] ChaMDE@MoDELS2008

  21. End-user Owner System Programmer Maintainer Tester Challenge 3. Multiviewpoint specifications Different stakeholders’ views Multiple aspects of a system: Consistency ChaMDE@MoDELS2008

  22. What is (in) a Multiviewpoint Specification? • This is the approach used by most EAFs • No correspondences between the viewpoint elements… … or trivially based on name matching • Others assume the existence of a global metamodel ChaMDE@MoDELS2008

  23. A global metamodel • Easier to manipulate from a theoretical point • Simplifies reasoning about consistency BUT… • The granularity and level of abstraction of the viewpoints can be arbitrarily different • The viewpoints may have very different formal semantics • Should it consist of the intersection or of the union of all viewpoints elements? • Both approaches have serious problems with extensibility and expressiveness (not to mention complexity of the second approach – think in the UML 2.0 metamodel). ChaMDE@MoDELS2008

  24. A global metamodel (i.e., Sauron’s approach to UML) The lord of the Metamodels (obviously, adapted) Three notations for the Structure modelers under the sky, Seven for the Behavior modelers in their halls of stone, Tree for Mortal Men doomed to die, One for the Designer of the Whole system on his dark throne In the Land of Mordor where the Shadows lie. One Metamodel to rule them all, One Metamodel to find them, One Metamodel to bring them all and in the darkness bind them In the Land of Mordor where the Shadows lie. ChaMDE@MoDELS2008

  25. Correspondences: Orthographic projections ChaMDE@MoDELS2008

  26. Multiviewpoint Specification ChaMDE@MoDELS2008

  27. Expressing correspondences • As Model Transformations • Possible if correspondences can be expressed as functions • Pairwise consistency can be formally studied • “One form of consistency involves a set of correspondence rules to steer a transformation from one language to another. Thus given a specification S1 in viewpoint language L1 and specification S2 in viewpoint language L2, a transformation T can be applied to S1 resulting in a new specification T(S1) in viewpoint language L2 which can be compared directly to S2 to check, for example, for behavioral compatibility between allegedly equivalent objects or configurations of objects” [RM-ODP, Part 3] • As Weaving Models • Possible if correspondences are just mappings ChaMDE@MoDELS2008

  28. Conclusions • MDE is more than Conceptual Modeling • MDE should also provide: • Development Methodologies • Predictable and Quantifiable Processes • Extensive Tool Support (Development, analysis, simulation, quality assurance, …) • Three (feasible) short-term challenges • Adding behavioral specs to DSLs • Adding NFP specs to DSLs • Coping with multi-viewpoint specs of systems • Everything with Tool support! ChaMDE@MoDELS2008

  29. Behaviour, Time and Viewpoint Consistency: Three Challenges for MDE José E. Rivera, José R. Romero, A. Vallecillo Universidad de Málaga Thanks! ChaMDE@MoDELS2008

More Related