On Premise Agent

When modernizing or migrating your enterprise application to Cloud, there may be instances, where you may have to connect to legacy data on-premise. CloudIO On-Premise Agent enables your cloud application to connect to your legacy data behind the firewall without having to open up any inbound ports.

On Premise Agent Architecture

How does it work?

The on-premise agent is an executable that can be configured to connect to your on-premise data that lives within your corporate firewall. Since the agent also runs within your corporate firewall it can access the data securely without having to expose any inbound port through the firewall.

You can set up an API Key & Secret for a service account on the cloud and set up the agent to use those credentials to make a secure WebSocket connection to the CloudIO Apps running on the cloud.

Note: If you CloudIO Apps are also running within your corporate network, then you don't need this On-Premise Agent.

When the user performs a transaction on the app running in the browser, it makes a call to the server running on the public cloud, which intern sends the request (Query, Update) down to the on-prem agent. The on-prem agent, upon receiving the request from the server connects to the on-prem database and performs the requested actions and returns the response to the server. The server will then forward the response to the user's browser.

Multiple instances of the Agent can be run to achieve high availability and scale. Each agent would connect to one of the many server nodes, the user application running on the browser also connects to one of the server nodes and Apache Kafka is used to instantly transport the request and response messages between the server nodes. With the use of Apache Kafka & WebSockets, the request from the browser to the cloud server and down to the agent and back to the cloud and then finally send back to the browser with very low latency (less than 100ms on average).