Why


The ADAS/AD challenge
Learn about the challenges of developing safe ADAS/AD vehicle functions


The AD Cycle - How it works
Learn about the AD Cycle works for efficient ADAS/AD development


Benefits of ETAS Deterministic Middleware Solution
See why the worlds top automotive OEMs choose us

Solution concepts


System representation concepts
Fundamental concepts to construct ADAS / AD systems


YAAA Modeling Concepts and Workflow
YAML-based concepts and workflows to model ADAS / AD systems


Deterministic Recompute and Virtual Drives
The 🔑 to development efficiency


Scheduling Mechanisms in ETAS DMS
Achieving optimal compromises between latency and predictability

Basics


Determinism
Gain deeper understanding on determinism in embedded real-time systems


Differences of μC and μP-based embedded systems
Why achieving deterministic behavior μP-based embedded systems is a challenge


Glossary
Understand the terms and abbreviations used in ETAS Deterministic Middleware Solution

Products


ETAS DMS Software Development Kit
Design, develop, build, debug and test ADAS/AD applications

ETAS DMS Middleware & Runtime Libraries
Safety-certified middleware and runtime libraries for series production

For whom


Software Architects
Architecture-as-code. Git-based collaboration. Auto-generated graphics.


Feature Developers
Reduced development complexity. Fast debugging with real vehicle data.

Services


Trainings
Get up to speed on developing with ETAS Deterministic Middleware Solution through our expert trainings


Consultancy
Want to know how to best make use of ETAS Deterministic Middleware Solution for your ADAS/AD platform or project?


Proof of Concept (PoC)
See one of your ADAS/AD vehicle functions developed with ETAS Deterministic Middleware Solution


Custom Project
Have a special need or require an extension to ETAS Deterministic Middleware Solution?

Support


FAQs
Get answers to frequently asked questions about ETAS Deterministic Middleware Solution


Contact
Get in touch with an expert to obtain support

Infos and news about ETAS Deterministic Middleware Solution


Blog
Get the latest news about ETAS Deterministic Middleware Solution


Newsletter
Get the latest news about ETAS Deterministic Middleware Solution delivered to your email inbox


Whitepapers
Read in-depth information, analysis, and insights from our experts

Hardware Accelerator Support (HWA)

Hardware Accelerator (HWA) Support via Direct Memory Connector (DMC)

Leveraging Hardware Accelerators (HWA) is paramount in the area of high-performance computing and machine learning. Our solution incorporates robust support for hardware accelerators through the Direct Memory Connector (DMC). This offers a seamless bridge between externally allocated memory and the ETAS DMS platform.

Key Components of HWA Support via DMC

  1. Variety of supported HWAs:

    DMC caters to diverse HWAs from various vendors, including general-purpose GPUs and specialized HWAs for Vision programming, Convolutional Neural Networks (CNNs), and Tensor processing.

  2. Middleware Frameworks:

    Integration with NVIDIA's middleware framework, NvSci, ensures efficient memory allocation and synchronization for devices like NVIDIA Xavier and Orin. This framework unifies host and device memory, incorporating CUDA, TensorRT, and NvMedia for optimal performance.

  3. NvSci Wrapper:

    CARMA, our middleware platform, provides a dedicated wrapper to facilitate direct access to NvSci-allocated memory for CPU processing. This eliminates the need for redundant data copies, enhancing overall efficiency.

Implementation of HWA support via DMC

  1. Handwritten HWA Business Logic:

    Specific HWA business logic, essential for high-performance tasks, is crafted manually by software developers. This tailored approach ensures optimal utilization of the underlying hardware capabilities.

  2. Gateway for Seamless Integration:

    The Gateway, comprising handwritten NvSci code, facilitates zero-copy data transfer from HWA to CPU activities. This seamless integration allows for efficient processing without the overhead of data duplication.

  3. NvSci Wrapper MWALA Plugin:

    The MWALA plugin implements the MWALA API for NvSci, enabling CPU activities to leverage custom NvSci code seamlessly. This integration, coupled with the NVSci Wrapper, ensures a smooth transition of HWA-acquired data into CPU processing workflows.

Benefits of HWA Support via DMC

  1. Elimination of Data Copy Overheads:

    The CARMA platform, with its NvSci Wrapper, eradicates the need for copying extensive payload data to CPU-accessible memory, enhancing overall processing speed and efficiency.

  2. Customizable Integration:

    The manual crafting of HWA business logic allows for tailored integration, ensuring that the unique capabilities of each HWA type and model are fully utilized.

  3. Middleware Abstraction:

    Leveraging the MWALA 2.0 abstraction on the ETAS DMS side ensures seamless middleware integration, making the system HWA-agnostic and versatile.

The HWA support via DMC empowers developers to harness the full potential of Hardware Accelerators, providing a streamlined and efficient solution for high-performance computing and machine learning applications.