LogoLogo
  • Introduction
  • Getting Started
  • SenseCrypt Server
    • Starting the server
    • Configuration
    • Licensing and Authorization
    • Using the Swagger Docs page
    • Authorization
    • Using a JWT Token for Mobile Authorization
    • Getting information about your license
    • SenseCrypt eID
      • Generating your first SensePrint eID QR
      • Generating a raw SensePrint
      • Decrypting a raw SensePrint
      • Parsing a SensePrint
    • SenseCrypt Face PKI
      • Configuration
      • Accessing your Root Certificate
      • Generate your first Face Certificate
      • Verifying a Face Certificate
      • Encrypting Data/Keys using a Face Certificate
      • Face Decrypting Data/Keys
      • Face Signing
      • Face Signature Verification
    • Accessing the server for testing
    • Liveness Image Requirements
  • SenseCrypt Mobile SDKs
    • Licensing
    • Authorization for Online SDKs
    • Liveness and Face Capture
    • Android SDK
    • iOS SDK
  • Conclusion
  • FAQ & Search
  • Appendix
    • Privacy Preserving Biometric Verifiability
    • Principles of Privacy Preserving Face Verification
    • Foundational vs Functional eID
Powered by GitBook
On this page
  • Pre-requisites
  • Best CPU Performance

Was this helpful?

Export as PDF

SenseCrypt Server

Installing and running the SenseCrypt Server

PreviousGetting StartedNextStarting the server

Was this helpful?

The first step in your eID journey is to run an instance of a SenseCrypt server. For your convenience, this is as easy as starting a Docker image via a provided shell script.

Pre-requisites

You need AMD / Intel bare metal machine / VM running Ubuntu 22.04 or Windows 11 server, and with Docker installed.

At least 8GB of free RAM and 4 CPU cores are needed for the server.

The server is optimized for CPUs and GPUs and container images are available for both CPU only (Linux/Windows) and CPU + GPU variants (Linux only).

For CPU + GPU varaints, the host machine must have NVIDIA drivers 550, 555 or 560 installed. Furthermore, the host machine must also have the NVIDIA container toolkit installed.

Best CPU Performance

For the best CPU performance, we recommend 4th generation Xeon hardware (Sapphire Rapids) with 8 cores or more.

While instance selection will highly depend on you own use-case, you can achieve 10x performance gains through BF16 inferencing offered by on these CPUs. The link shows some current instance types with 4th gen Xeon on various cloud providers.

Intel
NVIDIA
AMX
AMX