The gateway to federated learning

Governed, private, secure computational access to sensitive data for ML across boundaries. Control, manage and trace every computation.

Introducing the Apheris Compute Gateway

The Apheris Compute Gateway is an easily installable software package that lives alongside your data. Integrate and federate machine learning workloads built on sensitive data whilst retaining compliance with regulation. Data never has to move.

Learn more in our whitepaper on privacy and security.

Stay compliant

Leverage computational governance and control to allow data commercialization without ever giving direct access. Stay compliant with regulations and protect your IP.

No data pooling

The Compute Gateway lives alongside your data behind your firewall. Permit ML workloads running on your infrastructure and federate with others.

Integrate easily

ML engineers can work in their preferred ML frameworks by changing just a few lines of code. The Apheris Compute Gateway integrates easily with your environment and interfaces with existing systems and data sources.

All data governed, protected and secured

  • Data doesn’t move. Data is never directly accessible by third parties
  • Access to any Compute Gateway is secured via enterprise-grade features
  • Auditing and logging of user activities, actions, and executed computations
  • Communication to and across Compute Gateways is fully encrypted with TLS 1.2+. Data at rest is encrypted with AES-256

Control every aspect of a running algorithm

  • Use asset policies to precisely control who can do what on your data completely automated - adjust or revoke any time
  • Gain granular control at the computational level
  • Add human-in-the-loop approval for especially sensitive cases
  • Consult with Apheris privacy experts to help configure to your needs

Install, run, scale and federate ML across boundaries

  • Simply install the Compute Gateway alongside sensitive data either on-prem or in the cloud
  • Easy deployment via a lightweight Kubernetes distribution
  • Connect multiple Compute Gateways across organizational or regulatory boundaries while staying compliant with local requirements
  • Enable federated analytics and machine learning across Compute Gateways

Seamless integration into standard ML ecosystem

  • Allow users to run computations on Compute Gateways with minimal code change to standard machine learning frameworks
  • Built-in synthetic data module to allow safe data exploration for ML users without compromising data privacy
  • Easy discoverability of permitted datasets for ML users
  • State-of-the-art runtime performance
  • Apheris integrates with standard ML framework tooling

Certifiably secure

Securely working across organizational boundaries

Manufacturing

"With Apheris, we found the perfect partner for exploring privacy preserving data analytics and optimization along the value chain with our customers."

Christian Winkler, Director Global Infinergy Operations BASF

A gateway to data collaboration

Want to learn more about how Apheris can enable your path to governed, private, secure ML?