Case study

Connecting different VPC/VNets and extending on-premise resources to multiple clouds

Full stack development
PoC / MVP development

Business background

Clients icon

Client type

A networking software provider of SDN, WAN controllers, SD-WAN/LAN, private cloud, and DC provider
Goals Icon

Business goal

Expand the client's footprint in the enterprise market by enhancing the functionality of SDN to allow for an on-prem/cloud servers mixed connection

Our approach

  • Provide a complete team with experience in network engineering composed of an architect, project manager, and Python, Golang and UI developers
  • Prepare a Proof of Concept (PoC)
  • Ensure smooth communication between all stakeholders in a project involving all of the product’s important components
  • Deliver CI changes, automated tests and network configuration
  • Change the PoC to a full-size feature to be added to the customer’s SDN core product

Business benefits

  • Saved time: CodiLime’s team has a wide range of skills and was able to start working immediately.
  • Optimized spending: as a complete team was provided, the client did not have to expand its own team and incur considerable upfront investments.
  • Flexibility: with competence in multiple areas of network technical knowledge plus proficiency in software development and IT, CodiLime could efficiently explore the problem and quickly produce the optimal solution.

Technical highlights

  • The solution was based on deploying Multi Cloud gateway (MC-GW) nodes interconnecting different Virtual Private Cloud (VPC)/Virtual Networks (VNets) in the various clouds.
  • Secure tunneling was achieved by the simultaneous use of the OpenVPN, IPsec, and VXLAN.
  • Terraform was used to provision cloud resources on Microsoft Azure, Amazon Web Services (AWS) and Google Cloud Platform (GCP).
  • MC-GW also extends on-premise resources to the cloud.
  • Custom Ansible playbooks and Python scripts were used to deploy custom SD-WAN infrastructure and deploy the customer’s SDN solution.
  • Packer was used to improve the performance of consecutive provisionings.

Technology stack

Need support with your specific case?

For more information see our Privacy policy