AI Performance test on Cluster of Jetson Orin Nanos

Hi all,

I am testing the Jetson devices by running the Jetson Benchmarks (GitHub - NVIDIA-AI-IOT/jetson_benchmarks: Jetson Benchmark)

But seems the Github only supports single device.

Anyone has experience on running the benchmarks on multiple Jetson devices in parallel?

Thanks.

Hi,

The repository only supports the single device use case.

For the Jetson cluster, we recommended running different data on each device.
This can directly increase throughput without extra data transfer overhead.

Thanks.

Do you mean I have to distribute data to each device by myself?

Does Nvidia provide any documentation on running benchmark on jetson Cluster?

I am using Jetson Orin Nano 8GB

Hi,

No, our current benchmark is based on a single Jetson device.

Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.