---
title: Supported Connectivity Models
description: Check information about Integrations infrastructure
documentId: ipaas-infra-connectivity
locale: en-US
---
The data plane can be connected to public and private environments for:
Consumption of **public backend** for integration, as shown in diagram 1:
Consumption of **private backend**, as shown in diagram 2:
## Supported connectivity models
### Allow list (default)
**Positive points**
* Most recommended model due to practicality and resilience.
* Each data plane has at least 2 fixed outbound IPs.
These IPs can/should be used for *firewall* controls on the client side.
* Cost already embedded in standard offers.
**Limitations**
* It is not possible to access *backends* without externalization through a proxy or similar technology.
The diagram below represents connectivity via *allow list*:
### VPN (upon request)
**Positive points**
* Model with private access.
**Attention points**
* Shared responsibility.
* SLA is different for environments with VPN, given the high incidence of problems.
* Higher cost of configuration, maintenance, and troubleshooting.
**Limitations**
* Each data plane can be connected to up to 4 networks, limited to 8190 IPs.
* Currently BGP is not supported.
The diagram below represents connectivity via VPN:
### VPC peering (upon request)
**Positive points**
* Stability and resilience.
* Simplified configuration.
**Attention points**
* Offer available only to customers whose *backend* is also allocated on AWS.
**Limitations**
* Each data plane can be connected to up to 4 networks, limited to 8190 IPs.
The diagram below represents connectivity via *VPC peering*:
### Transit gateway (upon request)
**Positive points**
* Possibility to access *backends* through private link.
* No need to use VPN.
* Greater flexibility in communication between VPCs.
* Within AWS limits, essential points regarding transit gateway and connectivity are adjustable.
For more information, see [the documentation on quotas for transit gateways](https://docs.aws.amazon.com/pt_br/vpc/latest/tgw/transit-gateway-quotas.html).
**Attention points**
* Customer must share the AWS transit gateway with the Sensedia account.
* Need to create routes on both Sensedia and customer sides.
* AWS billing rate occurs on both sides (Customer and Sensedia), as AWS charges per VPC attached to the AWS transit gateway, and this happens on both sides.
**Limitations**
* Each data plane can receive up to 5 unique AWS transit gateway attachments.
* AWS transit gateway limits also apply.
The diagram below represents connection via transit gateway:
For more information, access the documentation about [connectivity with AWS transit gateway](/docs/integrations/infra-aws-transit-gateway).
### Direct connect (upon request)
**Positive points**
* Possibility to access *backends* through private link.
**Attention points**
* Cost.
* Shared responsibility model between Sensedia, customer, and link provider.
**Limitations**
Each data plane can be connected to up to 4 networks, limited to 8190 IPs.
The diagram below represents connection via *direct connect*:
Networks larger than 8190 hosts (/19) are not supported.
### PrivateLink (upon request)
**Positive points**
* Facilitates communication between components on AWS.
* Ensures private access with high resilience.
**Negative points**
* Requires exposure through NLB on the client side.
**Private DNS name**
* According to [AWS documentation](https://docs.aws.amazon.com/pt_br/vpc/latest/privatelink/manage-dns-names.html), it is possible to use a name with your own domain in the *endpoint service*, such as `.customer.com.br`.
This name is private, as AWS registers it in a local DNS zone for VPCs connected to the *endpoint service*.
* The advantage is being able to consolidate service exposure in a single name for different service consumers.
* Customers who wish to use the functionality need to enable and validate the *endpoint service* to use the selected name.
Once configured, the customer should open a ticket with Sensedia support, requesting activation of the private DNS name and informing the *endpoint service* in question.
**Limitations**
* Up to 4 VPC *endpoints* (enabled by *AWS PrivateLink*) are supported per data plane.
The diagram below represents connectivity using a VPC *endpoint* enabled by *AWS PrivateLink*:
Additionally, it is possible to use the same VPC *endpoint* associated with the same load balancer, with multiple ports and target groups.
The diagram below represents this model:
For more information, see [the official AWS documentation on PrivateLink](https://docs.aws.amazon.com/pt_br/whitepapers/latest/aws-vpc-connectivity-options/aws-privatelink.html).