For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).

Mojo function

shmem_init_thread_mpi

shmem_init_thread_mpi(ctx: DeviceContext, gpus_per_node: Int = -1)

Modular-specific init that enables initializing SHMEM on one GPU per thread.

Arguments: ctx: the DeviceContext to associate with this thread gpus_per_node: the number of devices participating on this node, by default this will use ctx.number_of_devices() to use all available GPUs.

Raises:

If SHMEM initialization fails.

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit