Appendix B: Third-Party Integrations¶
H2O Driverless AI integrates with a (continuously growing) number of third-party products. Please contact sales@h2o.ai to schedule a discussion with one of our Solution Engineers for more information.
If you are interested in a product not yet listed here, please ask us about it!
Instance Life-Cycle Management¶
The following products are able to manage (start and stop) Driverless AI instances themselves:
Name | Notes |
---|---|
BlueData | DAI runs in a BlueData container |
Domino | DAI runs in a Domino container |
IBM Spectrum Conductor | DAI runs in user mode via TAR SH distribution |
IBM Cloud Private (ICP) | Uses Kubernetes underneath; DAI runs in a docker container; requires HELM chart |
Kubernetes | DAI runs in as a long running service via Docker container |
Kubeflow | Abstraction of Kubernetes; allows additional monitoring and management of Kubernetes deployments. Click here for more information. |
Puddle (from H2O.ai) | Multi-tenant orchestration platform for DAI instances (not a third party, but listed here for completeness) |
SageMaker | Bring your own algorithm docker container |
API Clients¶
The following products have Driverless AI client API integrations:
Name | Notes |
---|---|
Alteryx | Allows users to interact with a remote DAI server from Alteryx Designer. Click here for more information. |
Cinchy | Data collaboration for the Enterprise, use MOJOs to enrich data and use Cinchy data network to train models |
Jupyter/Python | DAI Python API client library can be downloaded from the Web UI of a running instance |
KDB | Use KDB as a data source in Driverless AI for training |
RStudio/R | Under development, please ask for the DAI R API client library |
Scoring¶
The following products have Driverless AI scoring integrations:
Name | Notes |
---|---|
KDB | Call a MOJO to score streaming data from KDB Ticker Service |
ParallelM | Deploy and monitor MOJO models |
Qlik | Call a MOJO from a Qlik dashboard |
SageMaker | Host scoring-only docker image that uses a MOJO |
Trifacta | Call a MOJO as a UDF |
UiPath | Call a MOJO from within an RPA workflow |
Storage¶
Name | Notes |
---|---|
Network Appliance | A mounted expandable volume is convenient for the Driverless AI working (tmp) directory |
Data Sources¶
Please visit the section on Enabling Data Connectors for information about data sources supported by Driverless AI.