Workload orchestration between vehicle and cloud using DMS

on YouTube ↗11m

" summary

A PoC implements a container-based orchestration layer called Piccolo operator that schedules mixed-critical SDV workloads across vehicle, edge CDC nodes and cloud (Amazon Graviton) using the Abar platform. Two demos show concrete outcomes: auto scaling based on FPS raises DMS frame rate from ~10 to ~20+ by scaling to a CDC worker, and dynamic deployment of an eye-detection microservice pins safety-critical inference to the vehicle while offloading noncritical uploads to cloud.

AI-written summary · about

" chapters (8)

1. 0:00Workload orchestration between vehicle and cloud using DMS(Driver Monitoring System) use case in a mixed critical environment
2. 0:56How to achieve Mixed Critical Orchestration for SDV
3. 2:43Micro Service Architecture for SDV (Micro Service)
4. 4:35PICCOLO
5. 5:56PoC Concept
6. 7:38PoC System Architecture
7. 8:18Demo 1 – Auto Scaling based on FPS
8. 10:57Demo 2 – Add New Safety MSA feature

" related talks

• 08mMaximizing ROI Through Distributed Computing and Multi-CloudRavi Kumar Amaresam · Conf42 Machine Learning 2025
• 22mSteering into Tomorrow: Embracing Software-Defined Vehicle in an Open-Source Appro...Mahdi Shourabi · Automotive Linux Summit @ Open Source Summit Japan 2023
• 24mBeyond Tool Sprawl: An Event-Driven Approach to Dynamic Multi-Cloud OperationsEleni Grosdouli · ContainerDays Hamburg 2026
• 22mEmbracing Modern Workloads without Losing "IT"Yonggang Hu · MesosCon Asia 2016
• 01h 03mDeveloping microservices like a boss with Dapr and Azure Container AppsJakob Ehn · NDC Oslo 2023
• 20mEnhancing Workflow Reliability with DaprSiri Varma Vegiraju · Conf42 Cloud Native 2025
• 01h 02mA perfect match: Dapr & Azure Container AppsSander Molenkamp · NDC London 2023
• 50mChoreography vs Orchestration in serverless microservicesMete Atamel · Devoxx Ukraine 2021