- Proposed name for the project: Data Protection Service (Rescuer)
- Proposed name for the repository: Rescuer
- Rescuer is a project about data protection for OPNFV. Rescuer project would focus on developing use cases, requirement documents, and test cases. The use case would be submitted as blueprints to the upstream projects such as Karbor, Cinder, Swift, Ceph and so forth.Since Karbor is primarily designed for application data protection, especially in cross-site restore scenario, we will focus on Karbor as the main upstream project for Rescuer at the beginning and span to other upstream projects in later releases.
Figure 1:High level relationship between Rescuer and upstream projects
- Figure 1 shows the relationship between Rescuer and upstream projects. At first, we collect requirements about data protection from users or projects in OPNFV, then we will do gap analysis and develop use cases and test cases. In the next step the use cases will be submitted to upstream projects as blueprints and finally we provide the features for users in OPNFV in the form of documents.
Scenario 1: HA & Data protection in Storperf project
- In StorPerf, HA and data protection are very important when manager module deploys the Heat template and Agent VMs run performance tests.Consider these two scenarios:
- Heat saves the template in the local database, and all the templates could be gone if the VIM component goes down.
- If Agent VMs breaks down when running performance tests in storage allocated by Cinder, the manager can't get the report metrics and the configuration between VMs and storage would be lost.
- So how can HA and data protection solve those problems above? In HA scenario, OPNFV provides multiple VIM sites and compute nodes to handle single point of failure. And data protection service provides cross-site restore mechanism to protect resources from loss.
Scenario 2: Data protection in Barometer (SFQM)
- Barometer(SFQM) should connect to Ceilometer when this project works with VIM and Ceilometer will handle these metrics and stats gathered by SFQM. In this scenario, all the metrics data saved in Ceilometer would be gone if the VIM site breaks down. But right now there is no mechanism providing data protection service in OPNFV.
- Solve the problem of data loss or damage in OPNFV by providing upstream projects about data protection.
- Collect features about data protection, map them with use cases and finally provide the documents for users.
- Expand to DR in the later releases.
- Interoperability with Rescuer upstream projects for integration and testing.
- API Docs (TBD)
- Functional block description (TBD)
- Project has dependency on Karbor, Cinder, Swift, Freezer, Ceph and so forth. The features about data protection will be collected in the form of requirement documents.
Committers and Contributors:
- Hui Wang (Huawei): firstname.lastname@example.org
- Yuval Brik (Huawei): email@example.com
- Requirement documents released with OPNFV package.
- Developing use cases, requirement documents and test cases.
Proposed Release Schedule:
Release D plans:
- Focus on Karbor as the main upstream project for Rescuer at the beginning.
- Collect the requirements from users in OPNFV.
Release E plans:
- Gap analysis and developing use cases.
- Collect features and develop test cases in OPNFV.
Key Project Facts
Project Name: Data Protection Service (Rescuer)
Repo name: Rescuer
Lifecycle State: Proposal
Primary Contact: Hui Wang (firstname.lastname@example.org)
Project Lead: Hui Wang (email@example.com)
Jira Project Name: Data Protection Service
Jira Project Prefix: [Rescuer]
mailing list tag [Rescuer]
*Link to TSC approval: Example http://meetbot.opnfv.org/meetings/opnfv-meeting/2015/opnfv-meeting.2015-03-03-15.01.html*
Link to approval of additional submitters: Example http://meetbot.opnfv.org/meetings/opnfv-meeting/2015/opnfv-meeting.2015-03-03-15.01.html