Community Blog Integration of Alibaba Cloud as a Backend in OpenSDS and Object Storage for Multi-Cloud

Integration of Alibaba Cloud as a Backend in OpenSDS and Object Storage for Multi-Cloud

This blog embellishes the integration of Alibaba cloud as a back end in OpenSDS for multi-cloud.

*By Rupal Shirpurkar, Alibaba Cloud MVP and Business Head (Cloud BU - APAC) Click2Cloud Inc.

Before the integration of Alibaba, Gelato project of OpenSDS does not support object management on the Object Storage Service (OSS) of Alibaba Cloud. The Gelato project is developed by the SODA Foundation and is typically used for multi-cloud projects, supporting object operation performance, lifecycle management, and object management for different cloud vendors.

Alibaba Cloud as a back end in OpenSDS multi-cloud project integration helps in object operation, lifecycle management. Users can perform in-cloud and cross-cloud migration for Alibaba cloud using the Gelato project for Alibaba cloud OSS service.

OpenSDS is an open-source community working to address storage integration challenges, particularly in scale-out cloud-native environments with heterogeneous storage platforms.

The image below illustrates the architecture of the integration of Alibaba cloud in OpenSDS/ multi-cloud.

Fig: Architecture of Integration of Alibaba cloud in OpenSDS/multi-cloud

Alibaba Cloud Object Storage Service (OSS) helps users to conduct operations related to bucket and object storage in simple ways. Alibaba OSS adapter helps the user with:

  • Adding Backend for Alibaba OSS
  • Creating OpenSDS Bucket on Alibaba OSS
  • Deleting Bucket from Alibaba OSS
  • Support for SSE
  • Uploading Object on Alibaba OSS (PUT Object)
  • Downloading Object from Alibaba OSS (GET Object)
  • Support for Multi-part Upload
  • Deleting Object from Alibaba OSS
  • In-cloud lifecycle management on Alibaba OSS (Standard, IA, Archive)
  • Cross-cloud lifecycle management with Gelato supported backends (i.e. AWS, Azure & Huawei)

For a multi-cloud data engine, the team sketched the architecture diagram that presents the ability to migrate data across multiple clouds. The multi-cloud architecture diagram defines the concept of multi-cloud in OpenSDS clearly. Also, the user will be able to migrate its bucket from any OpenSDS supported public cloud or local storage to Alibaba Cloud.

Let us see some of the significance of the architecture diagram above:

  1. API Gateway: At the topmost, an API Gateway is shown, which takes client request and provides a response with backend services collection.
  2. S3 Service: It contains the S3 controller which communicates with the database and object storage backends on-premise or in the cloud, handles S3 API requests, returns results to the caller. To convert the location of an object passed in from API into the actual location on the object storage S3 service uses Location Mapper.
  3. Backend Service: This service communicates with the backend on-premise or in the cloud to confirm the information about the bucket. Also, backend service manages object storage backends on-premise or in the cloud.
  4. Dataflow service: Dataflow service is used for data migration.
  5. Data Mover: This is using to move the data from one location to another location.
  6. Database: Database is used to save the information that is present on other resources of a multi-cloud data engine.
  7. Backend Adapter: To support and communicates with backend management like AWS, Backend Adapter is used.

Features of Alibaba Cloud Integration as a Backend in OpenSDS/Multi-cloud.

The integration of Alibaba cloud in OpenSDS/multi-cloud architecture offers number of reasons to deploy into it which often results in reducing reliance on a single vendor, increasing flexibility, cost-efficiency, adherence to local policies, and a lot more.

Users use OpenSDS multi-cloud data engine to move data to a location near to where it is needed for convenience. Data can be moved to different locations depending on usage, age, access frequency.

Alibaba integration with the OpenSDS data engine allows data to be available and accessible all the time. In the case where one cloud is down, the user can still have access to data in another cloud that is up and running.

A multi-cloud data engine provides policy-based data mobility across public and private clouds. These public and private clouds can be used in the following cases:

  • Users may want to upload some datasets to the public cloud for data analysis and delete the data after the work is complete.
  • Users can leverage a multi-cloud data engine to move data to the public cloud for high CPU computing when that is required for a period.
  • Frequently accessed data can be stored on-premise. Data not accessed for a long time can be stored in archival storage in public clouds.


Without this integration, users were not able to perform object related operations for Alibaba Object Storage Service on Multi Cloud Project of SODA (i.e. Gelato Project). Alibaba Cloud has committed to extend the OpenSDS capabilities for Data Storage and Migration for a multi-cloud design and multi-cloud providers. Integration with Alibaba cloud as backend in OpenSDS leads to a solution to many cloud storage vendors to come up with Object Storage Service (OSS) issues.

0 0 0
Share on


5 posts | 0 followers

You may also like



5 posts | 0 followers

Related Products