Hyperledger Fabric on Azure Kubernetes Service Marketplace template

Data di pubblicazione: 30 gennaio, 2020

Program Manager

Customers exploring blockchain for their applications and solutions typically start with a prototype or proof of concept effort with a blockchain technology before they get to build, pilot, and production rollout. During the latter stages, apart from the ease of deployment, there is an expectation of flexibility in the configuration in terms of the number of blockchain members in the consortium, size and number of nodes and ease in management post-deployment.

We are sharing the release of a new Hyperledger Fabric on Azure Kubernetes Service marketplace template in preview. Any user with minimal knowledge of Azure or Hyperledger Fabric can now set up a blockchain consortium on Azure using this solution template by providing few basic input parameters.

This template helps the customers to deploy Hyperledger Fabric (HLF) network on Azure Kubernetes Service (AKS) clusters in a modular manner, that meets the much-required customization with regard to the choice of Microsoft Azure Virtual Machine series, number of nodes, fault-tolerance, etc. Azure Kubernetes Service provides enterprise-grade security and governance, making the deployment and management of containerized application easy. Customers anticipate leveraging the native Kubernetes tools for the management plane operations of the infrastructure and call Hyperledger Fabric APIs or Hyperledger Fabric client software development kit for the data plane workflows.

The template has various configurable parameters that make it suitable for production-grade deployment of Hyperledger Fabric network components.

Top features of Hyperledger Fabric on Azure Kubernetes Service template are:

  • Supports deployment of Hyperledger Fabric version 1.4.4 (LTS).
  • Supports deployment of orderer organization and peer nodes with the option to configure the number of nodes.
  • Supports Fabric Certificate Authority (CA) with self-signed certificates by default, and an option to upload organization-specific root certificates to initialize the Fabric CA.
  • Supports running of LevelDb and CouchDB for world state database on peer nodes.
  • Ordering service runs highly available RAFT based consensus algorithm, with an option to choose 3,5, or 7 nodes.  
  • Supports ways to configure in terms of number and size of the nodes of Azure Kubernetes Clusters.
  • Public IP exposed for each AKS cluster deployed for networking with other organizations
  • Enables you to jump start with building your network sample scripts to help post-deployment steps such as create workflows of consortiums and channels, adding peer nodes to the channel, etc.
  • Node.js application sample to support running a few native Hyperledger Fabric APIs such as new user identity generation, running custom chain code, etc.

To know more about how to get started with deploying Hyperledger Fabric network components, refer to the documentation.

What customers are saying 

“At KPMG we are building an enterprise-grade track and trace platform on Hyperledger Fabric called KPMG Origins. When we started the project, there was no product offered by cloud providers to deploy Hyperledger Fabric on Kubernetes, so we had to build Hyperledger deployment templates from scratch. The process to build and maintain this complex Blockchain deployment infrastructure is a resource and time-consuming process. Given the size of the team, our goal is to invest time in building applications on Blockchain rather than spending time on setting up the infrastructure. After trying out the Hypereldger Fabric on Azure Kubernetes Service marketplace offering, it has simplified our DevOps pipelines to spin up a consortium of Blockchain nodes and we no longer need to maintain our custom templates.” —Gokulnath Reddy, Blockchain Tech Lead – KPMG Origins

What's coming next

  • Microsoft Visual Studio code extension support for Azure Hyperledger Fabric instances

What more do we have for you? The template and consortium sample scripts are open-sourced in the GitHub repo, so the community can leverage to build their customized versions.