https://github.com/ow2-proactive/prestocloud-tosca-interpreter
A Java parser for TOSCA Simple Profile in YAML Version 1.2
https://github.com/ow2-proactive/prestocloud-tosca-interpreter
Last synced: 2 months ago
JSON representation
A Java parser for TOSCA Simple Profile in YAML Version 1.2
- Host: GitHub
- URL: https://github.com/ow2-proactive/prestocloud-tosca-interpreter
- Owner: ow2-proactive
- License: apache-2.0
- Created: 2021-03-08T11:40:12.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2021-03-10T19:34:05.000Z (about 4 years ago)
- Last Synced: 2025-01-25T17:15:42.581Z (4 months ago)
- Language: Java
- Size: 966 KB
- Stars: 0
- Watchers: 18
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# TOSCA Interpreter
This repository contains the TOSCA interpreter. This component is in charge of interpreting both type-level TOSCA and instance-level TOSCA:
- Type-level TOSCA interpretation
- Parsing the type-level TOSCA file describing an application to be parsed
- Assessing computing resources availability
- Programming a Btrplace model
- Executing btrplace solving
- Interpreting a reconfiguration plan matching with TOSCA specification and resources availability
- Determining hourly operation cost of the new deployment
- Generating instance level TOSCA
- Instance-level TOSCA interpretation
- Parsing the instance-level TOSCA file describing an application instance
- Counting edge and cloud instance per fragment
- Restitute it in a format compatible with the [Application Fragmentation & Deployment Recommender](https://gitlab.com/prestocloud-project/application-fragmentation-deployment-recommender).
In the context of the PrEstoCloud architecture, this component implements both:
- *The ADIAM*: It interprets the input from the meta-management layout, and determine which management operation should be enacted, and communicate back the current status of deployed resource to the meta-management layer.
- *The APSC*: It performs the solving of the fragment placement problem. This feature is supported by the BtrPlace library, provided by CNRS.
## Building the project
The project uses maven to retrieve the build dependencies, perform the compilation and the package the outcome.
1. Perform the compilation with `$ mvn package -Dmaven.test.skip=true`
2. Retrieve the JAR executable from `target/` directory.
## Testing the project in a local environment
The component can be tested offline of the whole ADIAM platform with the following command:
### Type-level TOSCA interpreter
```
java -jar target/prestocloud-tosca-1.0.0-SNAPSHOT.jar type-level-interpreter
```
The argument are the followings:
- *tosca_repository* : Refering the TOSCA definition file necessary to tosca file parsing. A valid directory containing valid definition has to be specified. `src/main/resources/repository/` is provided as an example in the repository.
- *tosca_resource*: Refering to the directory containing tosca file defining cloud and edge resource to be schedulled. A valid file containing cloud resource and edge resource definition has to be mandatorily specified. `src/test/resources/prestocloud/resources/` is provided as an example.
- *type_level_tosca_file*: Identifying the tosca file to be parsed and interpreted into an ADIAM's reconfiguration plan. This parameter must refer to a valid TOSCA file.
- *reconfiguration_deployment_file*: Pointing to the ADIAM reconfiguration file to be created. This file is to be consumed later by the [ADIAM main workflow](https://gitlab.com/prestocloud-project/adiam-workflows).
- *instance_level_tosca_file*: Providing the instance level TOSCA file to be produced when executed in teh ADIAM environment (i.e. the variable environment `variables_CLOUD_LIST` is set). The file has not to necessarily exist. The produced file contains macro to be substituted by the ADIAM before being able to be interpreted.
- *mapping*: Containing the scheme of an already existing deployment. Can refer to a non-existing file for initial deployment.
- *edge_status_file*: Mandatorily referring to a file containing an output of the [ProActive Resources Manager](https://github.com/ow2-proactive/scheduling) API call to topology endpoint. An example file should have the following content:
```
[
{
"timeStamp": 0,
"counter": 1,
"firstCounter": 1,
"nodeSourceName": "edgeDevice",
"nodeSourceDescription": "Infrastructure: Default Infrastructure Manager, Policy: Static Policy user access type [ALL], provider access type [ME]",
"additionalInformation": {},
"nodeSourceAdmin": "rm",
"nodeSourceStatus": "deployed",
"sourceDescription": "Infrastructure: Default Infrastructure Manager, Policy: Static Policy user access type [ALL], provider access type [ME]",
"sourceName": "edgeDevice",
"key": "edgeDevice",
"rmurl": null,
"eventType": null,
"timeStampFormatted": "1/1/70 1:00 AM"
}
]
```
Command usage example:
```
java -jar target/prestocloud-tosca-1.0.0-SNAPSHOT.jar src/main/resources/repository/ src/test/resources/prestocloud/resources/ simple_tosca_deployment_php_mariadb.yaml output.json instance_level_tosca.yml mapping.json status.sample.json
```
### Instance-level TOSCA interpreter
```
java -jar /target/prestocloud-tosca-1.0.0-SNAPSHOT.jar instance-level-interpreter
```
The arguments are the following:
- *tosca_repository*: Referring the TOSCA definition file necessary to tosca file parsing. A valid directory containing valid definition has to be specified. `src/main/resources/repository/` is provided as an example in the repository.
- *instance_level_tosca_file*: Identifying the Instance-TOSCA file to be interpreted and parsed. This parameter must point to a valid TOSCA file.
- *metamanagement_output_file*: File to be produced to be communicated to the [Application Fragmentation & Deployment Recommender](https://gitlab.com/prestocloud-project/application-fragmentation-deployment-recommender).
Command usage example
```
java -jar /target/prestocloud-tosca-1.0.0-SNAPSHOT.jar instance-level-interpreter src/main/resources/repository/ instanceleveltosca.yml output.txt
```
## Installing in the ADIAM platform
This interpreter is designed to be installed inside the ADIAM platform, and called from the [main workflow](https://gitlab.com/prestocloud-project/adiam-workflows).
To proceed so, the package has to be uploaded on the server operating the ProActive instance loaded with the ADIAM workflow, to the following path:
```
/home/ubuntu/tosca-parser/prestocloud_tosca_parser.jar
```