Heterogeneous Computing
A computing architecture that employs different processor types—such as CPU, GPU, and NPU—working in coordination. By assigning tasks to the most suitable processing unit, this approach satisfies both intelligence and real-time performance requirements within constrained power budgets and physical form factors.
Timeline (last 90 days)
In Intel's physical AI solution, the CPU, GPU, and NPU are each assigned different tasks such as AI inference, environmental perception, task planning, and motion control.
Heterogeneous computing refers to combining processing units of different architectures—such as CPUs, GPUs, and dedicated accelerators—within the same system to dynamically allocate computing resources based on task characteristics.
The concept of heterogeneous computing can be traced back to GPGPU research in the 1990s, and only became mainstream starting in 2007 when NVIDIA introduced the CUDA programming model
All Facts (3)
In Intel's physical AI solution, the CPU, GPU, and NPU are each assigned different tasks such as AI inference, environmental perception, task planning, and motion control.
50%UnverifiedHeterogeneous computing refers to combining processing units of different architectures—such as CPUs, GPUs, and dedicated accelerators—within the same system to dynamically allocate computing resources based on task characteristics.
50%UnverifiedThe concept of heterogeneous computing can be traced back to GPGPU research in the 1990s, and only became mainstream starting in 2007 when NVIDIA introduced the CUDA programming model
50%