Presentation is loading. Please wait.

Presentation is loading. Please wait.

ONAP Integration Through Information and Data Modeling

Similar presentations


Presentation on theme: "ONAP Integration Through Information and Data Modeling"— Presentation transcript:

1 ONAP Integration Through Information and Data Modeling
12 December 2017 Andy Mayer, Ph.D.

2 ONAP Information Integration GOAL
We need a cohesive way to have a shared view of information across ONAP (and on ONAP External APIs, e.g., BSS/OSS, Partner Providers, Multi-VIM, third party clouds, etc.) that can be used for or be input into a model driven process whereby the cost of delivering platform functionality is drastically reduced and the time to delivery is dramatically decreased.

3 ONAP Information Integration Challenges
Currently there are a number of information and data modeling efforts of varying similarity covering overlapping problem spaces These models often apply different models for the same concept or the same class name for different concepts The modeling patterns differ. E.g., some are inheritance based, others are composition based, etc. © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

4 Model Driven ONAP Traditionally, introducing new elements, new products, and new services require long software development cycles A key objective is to eliminate these long cycles by: Standards to establish common ways provide lifecycle management of VFs in support of operational integration Abstractions to support new types of Services and VFs with minimal changes Generic ways to monitor the infrastructure Standard functions with fixed inputs and outputs that are tunable via metadata Models and processes that are designed, distributed, and interpreted by the execution environment

5 The Model Driven Paradigm
Models Execution Engines Behaviors Resource definition Service composition Product models Lifecycle models (Operational Integration) Processes and policies Optimization rules Software images Orchestration Policy execution Data movement Data storage Coded algorithms Control Standard data models Real-time service delivery Elastic scalability Self-optimizing Self-healing

6 Inter-relationship between Services and Network Functions
Design Service Instantiation Designed Service Descriptor Instantiated Service Service Lifecycle Management VNF On-Boarding VNF Instantiation VNF Package VNF Descriptor Instantiated VNF Resource Lifecycle Management

7 Role of External APIs in Model Driven ONAP The Story
6. BSS/OSS builds Products based on Service Descriptors 7. Based on a Product Order, BSS/OSS requests Service instantiation based on Service Descriptor 8 . SO uses Service Parameters from request and corresponding Service Recipe to instantiate the Service 5. BSS/OSS retrieves Service Descriptors from the Catalog 1. VNF Package with VNFD is On-Boarded into SDC Run-time Design-time (SDC) Policy Creation & Validation Analytic Application Design Recipe / Engineering Rules & Policy Distribution VNF SDK ONAP Operations Management (OOM) External Data Movement & APIs Common Service Micro Services Bus ONAP Portal – Design Studio & Dashboard Infrastructure OpenStack Azure VMware RackSpace ...... Resource Onboarding Service & Product Design DMaaP AAF BSS/OSS Catalog Products Resources Eng. Rules Services Policies Analytics Testing & Certification Multi-VIM Open Stack AWS Model Driven Adaption Layer Service Logic Interpreter Config Database Yang Netconf CLI Life Cycle Mgmt DGs L0-3 Network Controller Logging VNF 1 VNF 2 VNF n PNF Config Database Standard & sVNFM Adaption Layer Vendor B Vendor A Service Logic Interpreter Life Cycle Mgmt func. Chef Ansible L4-7 Generic VNF Controller gVNF-Controller (prev. App-C) sVNFM EMS Modeling (specs & Utilities) SDN-C VNF x Data Collection Layer DCAE Data Distribution CDAP Holmes Orchestration Framework Service Orchestration Resource Orchestration Analytic µServices Runtime Policy Execution Engine Resource / Service Topology Active / Available Entitlements A&AI Lab Integration & Certification 3rd Party Network controller Policy ESR 2. SDC designs a service using the VNFD. Design includes Service Descriptor and Service Recipes 4. Service Descriptors are made available via the Catalog 3. Service Recipes are distributed to ONAP components (e.g. SO)

8 Information Domains in ONAP
Governance/Processes Policy/Rule/Constraints, SLA/OLA Definition, Asset/Capacity Management, Configuration/Change/Incident Management, Catalog Management Service Design Network function on-boarding Service definition, including Service blueprints Service Access User/Tenant Management, Identity Management, RBAC Entitlement Control Services Policy/Rule/Constraints, SLA/OLA Definition Asset/Capacity Management, Configuration/Change/Incident Management Services Components Decomposition/Instantiation/Assurance/Remediation Service Level Optimization, QoS Optimization, Policy Enforcement, Logical Resource Resource Matchmaking/Optimization Resource Reservations/Allocations, Installation/Configuration Real Infrastructure Provider Registration, Infrastructure Discovery/Monitoring Capacity Optimization Information Domains in ONAP Governance/Processes Services (What’s offered; Catalog) Service Components (Service Function Topology; Network Functions) “Logical” Resource (VNFs; VNF Forwarding Graphs) External APIs “Real” Infrastructure (Physical/Virtual/Cloud – On/Off-Premise) Service Access Service Design Micro-service Bus /Integration Backplane

9 So…. What’s an Information Model?
Information Models define managed objects at a conceptual level, independent of any specific implementations or protocols used to transport the data A semantically meaningful representation of concepts (classes of entities) Described in an implementation independent manner Class characteristics (attributes) Relationships among classes Operations (actions) performed on classes by actors Useful to describe the managed environment, and as a guide to the information and behavior that will be described and coded in the Data Models. May be represented in Unified Modeling Language (UML) © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

10 Benefits of Information Modeling
Common Terminology and Definitions: The Information Model captures common terms, semantics, and definitions for shared information. Stewardship of Information: Identifies the domain of information to be managed and controlled by a stewarding system and the contexts associated with each; Management of Information Integrity and Redundancy: Identify where information is produced and made available. An Integrated Information Strategy should identify the processes for maintaining the integrity of Information; Interface / Microservice Characterization: Integrated Information Model will be used to describe and characterize interfaces / microservices and separations of concern; Consistent Methodology, Tools, and Processes: Define and establish the methodology, processes, and supporting tools; © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

11 Information Model Guiding Principles
Abstractions and Constructs: described in terms of information categories and high level information classes including properties (e.g., attributes, states, associations, etc.) Focus on Shared Information that is exchanged among the interoperable components within ONAP and with external actors. Document properties behavior and semantics in Integrated Information Model Define a common platform independent representation of shared information, while providing the means to extend the model in a semantically rich fashion Ensure that ONAP functionality and information is developed in a logically consistent fashion, allowing rapid integration Enablement of Model Driven Architecture and Design © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

12 What about Data Models? Data Models define managed objects at a lower level of abstraction than Information Models. Data Models include implementation and may include protocol specific details. A Data Model formally describes the structure and the semantics of all the data that are (a) stored in data management systems, and/or (b) accessed and manipulated by data-centric applications across APIs. The information exchanged across an API is described within a data model that is specified in a data modeling language, for example YANG or YAML An API also defines the encoding format that is used to encode data into a representation and format that can be exchanged across the interface according to the structure described by the data model, and the protocol that is used to carry the encoded interface data (e.g. NETCONF, RESTCONF or REST/HTTP). © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

13 IISOMI (Informal Inter- SDO Open Model Initiative)
multi SDO UML to DS Mapping Guidelines IISOMI 531: UML to YANG Mapping Guidelines IISOMI 515: Papyrus Guidelines UML | YANG IISOMI 514: UML Modeling Guidelines Pruning/Re-factoring Process Guidelines UML | XML IISOMI (Informal Inter- SDO Open Model Initiative) Guidelines UML | JSON guide guide guide UML | TOSCA guide Common Information Model pruning re-factoring a mapping Tooling YANG xx b pruning re-factoring mapping Technology specific UML Models c xx TR-512: Core Network (Forwarding, Topology, Termination, Foundation, …) pruning re-factoring mapping App specific z TOSCA xx Core Fragment Specific Fragments Purpose-specific IMs Interface-specific Data Schemas Interface-specific Encodings guide guide guide guide guide guide guide TR-513: Common Information Model Overview (structure, development process) IISOMI / Bernd Zeuner (DT)

14 The ONAP Pieces ONAP Common Information Model ONAP Purpose-
Specific Information Model ONAP Data Dictionary ONAP Component Design & Development © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

15 Data Dictionary A comprehensive list of the possible types of attributes, since attribute types are the core re-usable components of a data model; The list of individual attributes and their types in the data model, since attributes can be reused in different entities or relationships; The naming convention used for types, attributes, entities and relationships, to ensure a common language and help avoid ambiguity; and Examples of usage of types (in entities or relationships, even though entities and relationships are not part of the data dictionary itself), naming conventions, etc. © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

16 General Integration Methodology
Identify common Information Constructs from an ONAP wide perspective using Use Case interactions, microservice & interface definitions as input. Continuous Integration with appropriate standards and relevant ONAP efforts related to architecture, information modeling, and microservice & interfaces Use UML tools (e.g., Eclipse Papyrus) to capture living ONAP Integrated Information Model and Use Cases Revise descriptions by adding additional details as they are discovered Identify main consumers and producers / stewards of the Shared Information from a functional standpoint Share and gather feedback of model content with key stakeholder project Apply and Map model when designing and developing data models for APIs and microservices © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

17 ONAP Information Modeling Role
Guide the development ONAP Abstractions and Constructs, and form an Integrated Information Strategy and Common Model: Provide a coherent view of shared information Identify common abstractions and constructs across ONAP Capture common terms and definitions with respect to shared information Help identify unplanned redundancy Facilitate interactions for common interfaces (and microservices) Identify the primary authority (or Steward) for each piece of shared information from a functional perspective Form a common Integrated ONAP Information Model to be used as an integration resource (Federated approach) © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.

18 ONAP Information Model Topics Timeline
Draft Release, October 31, 2017: Network Updates: Connection Points; Virtual Link Descriptors; VNF Forwarding Graph Descriptor VNF Descriptor Model (w/ initial Deployment Flavors) Initial Service Descriptor / Service Component Draft Release, Mid-December, 2017: Multi-VIM Concepts Element Groups Refinement of Deployment Flavors Service Descriptor Refinement VNF Descriptor Refinement (e.g., Extensions, Event Descriptors, Healthcheck, etc.) Virtual Resource Elements Service and Resource Lifecycle Management Concepts 2018 Backlog: License Descriptor and Usage Service and Resource State (initial) Service Path as a Service Policy Model Assurance (including control loop management) Address Space Management © 2016 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.


Download ppt "ONAP Integration Through Information and Data Modeling"

Similar presentations


Ads by Google