tft_notify_controller_on_global_rank

This API is called by MindCluster to notify MindIO TFT of the global fault NPU information.

mindio_ttp.controller_ttp.tft_notify_controller_on_global_rank(fault_ranks: dict,time:int=1)

Parameter	Mandatory/Optional	Description	Value
fault_ranks	Mandatory	Information about the faulty NPU.	<int key, int errorType> dictionary: key: rank ID of the faulty NPU errorType: fault type 0: UCE 1: non-UCE fault
time	Optional	Maximum time for interacting with the MindCluster repair policy, which is determined based on the environment variable.	The value is an integer in the range [1, 3600], with a default of 1 (unit: seconds).

Parameter

Mandatory/Optional

Description

Value

fault_ranks

Mandatory

Information about the faulty NPU.

<int key, int errorType> dictionary:

time

Optional

Maximum time for interacting with the MindCluster repair policy, which is determined based on the environment variable.

The value is an integer in the range [1, 3600], with a default of 1 (unit: seconds).

Parent topic: API Reference