Distillation of requirements (Schema Service)

1.

Provide schema per partes

@William Reehil/AT&T

For example, say Sparky just needs to know the attributes of the x node, it should not have to load the entire configuration file in XML format to find out that information. Ideally it should just be able to ask the schema service "What are the attributes of the x node in JSON format?"

1. Sparky

2. GraphGraph Chatbot (Future)

3. GraphGraph UI (Future)

4. Any mS needing schema information without the need to process whole OXM (lightweight UIs & mS)

5. Cacher

Tier 2

Operations Users

Software Engineers

AAI Onboarders (Devs)

2.

Schema modifications hub

@William Reehil  /AT&T

Then taking this a step further we would like to leverage this service to act as the central hub of schema modifications

AAI platform & development architects to support multiple instances of the AAI platform simultaneously with various schemas, giving a centralized hub of schema management

Software Engineers

NARAD & ECOMP(AT&T Business Use Cases)

3.

Adding schemas at runtime

@William Reehil /AT&T

ability to consume new schema dynamically, without downtime

AAI platform & development architects and Tier 2 to mitigate downtime of the platform when processing new SDC models for clients to interact with.

Every AAI client

(20+ clients)

Tier 2

4.

Interface support

@William Reehil /AT&T

interface to validate proposed schema changes

AAI platform & development architects to ensure a valid schema is used, so as to not to not disrupt the current production graph when consuming updates. This effort would help ensure lower rejections from our clients and higher reliability of the platform, so Tier 2 can manage the platform more efficiently.

Every AAI client

(20+ clients)

Tier 2

5.

Input format support

@William Reehil /AT&T

Ability to import SDC TOSCA/CSAR schemas as well as

ability to provide the schema via flexible document formats (OXM, TOSCA etc.)

AAI platform & development architects to centralize the authoring/modification of the model across all applications.

@Chesla Wechsler

6.

Dynamic schema updates

@CT Paterson /Amdocs

@William Reehil and @Manisha Aggarwal /AT&T

@Keong Lim / Huawei

Ability do dynamically update schemas in a running schema service without downtime. Changes to the schema are persisted "forever" (assuming even after redeploy).

1. According to @CT Paterson this functionality would be appreciated by an unknown Amdocs customer

2. According to @William Reehil this step is a preparation for "SDC model → A&AI target architecture"

3. According to @Keong Lim it would make development of CCVPN use case easier as changes can be made dynamically

ONAP AAI Adopters ( ie. Amdocs Business Partners)

7.

Support non-standard schemas

@CT Paterson /Amdocs

Support standard and non-standard schemas, as some customers may have different schemas altogether. In terms of usage is related to the requirements Dynamic schema updates

According to @CT Paterson this functionality would be appreciated by an unknown Amdocs customer

NARAD

Amdocs Business Partners

ECOMP