Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Feature List

  • By   (AB meeting in )
  • Virtualization on  
  • Containerziation on  

Top Level Features:

Feature 1: Define the requirements and use cases of Software-defined Vehicle with AGL, focusing on the environment parity (including both application-level and operating-system-level)

  • Persona
    → define users/developers/testers/... (user journey)
    → infotainment/telematics
  • Define the meaning of "Workload" and "Workload Orchestration"
    • single SoC workload orchestration
    • multi-SoC (zonal architecture) workload orchestration
    • from vehicle to cloud (bi-directional)

Feature 2: Implement PoC with step-by-step MVPs containing key enabling technologies (Device Virtualization, Workload Orchestration and etc) for SDV to fulfill the requirements and use cases defined.

Info
  • Define the requirements of AGL for Software-defined Vehicle
    • Virtualization (hypervisor, device virtualization "VirtIO)
    • Conternization (application container and orchestration)
    • Cloud-native
  • Clearly define the use case that the group as a whole has consesus and develop a prototype that community members can work together (link with IC, IVI-EG):  A pure OSS solution using both virtualization and conternization

Technical Features:

Feature 3: Further extend and enhance VirtIO as a common AGL device virtualizaiton framework to decouple AGL software from underlying SW&HW platform on both physcial ECU in edge and virtual ECU in cloud.

Feature 4: Integrate Display Virtualization solution "Unified HMI" into AGL UCB to actualize cross-ECU/processor device virtualization
Feature 5: Develop a prototype of workload orchestration & deployment starting from single SoC envionrment (workload examples: container, thin VM, posix process and etc.)

Info
  • Virtualization
    • VirtIO in HV
    • VirtIO in non-HV
    • VirtIO across Multi-ECU
    • VirtIO in Cloud-Native
  • Conternization & Orchestration
    • Determine Container Type suitable for automotive use case:
      • LXC
      • docker
      • container-d
      • podman
      • ???
    • Evaluate the usage of application container inside AGL for IVI/IC use cases
      • implementaiton in SOAFEE is lacking graphic perspectives which has higher weight in AGL (SOAFEE is targeting more for ADAS/AD features)
    • Find/develop an OSS container engine and orchestrator (k3s?) fitting for AGL use case
  • Cloudification
    • Cloud-native Environment
    • Cloud service
    • CICD cloud infrastructure