[object Object]

[object Object][object Object]undefined

Note: When using this API, ensure that the driver firmware package and CANN package are in the 8.0.RC2 version or later. Otherwise, an error, such as BUS ERROR, will be reported.

[object Object]

• API function: Integrates AllGather communication and MatMul compute.
• Formula:$$output=AllGather(x1)@x2+bias$$ $$gatherOut=AllGather(x1)$$

[object Object]

Each operator has calls. First, [object Object] is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, [object Object] is called to perform computation.

[object Object]

• Parameters

  [object Object]
  • [object Object]Atlas A3 training products/Atlas A3 inference products[object Object], [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]:
    • bias: The scenario where the input is not 0 is not supported.
  • Ascend 950PR/Ascend 950DT:
    • bias: The scenario where the input is not 0 is supported.

• Returns

  [object Object]: status code. For details, see .

  The first-phase API implements input parameter verification. The following errors may be thrown.

  [object Object]

[object Object]

• Parameters

  [object Object]

• Returns

  aclnnStatus status code. For details, see .

[object Object]

• Deterministic computing:

  • [object Object] defaults to a deterministic implementation.

• Input [object Object] is 2D (m, k). x2 must be 2-dimensional, and its shape is (k, n). The axis meets the input parameter requirements of the MatMul operator. The k axis is equal, and the value range of the k axis is [256, 65535).

• [object Object] and [object Object] support empty tensor scenarios. m and n can be empty, k cannot be empty, and the following conditions must be met:

  • m is empty, k is not empty, and n is not empty.
  • If m is not empty, k is not empty, and n is empty.
  • If m is empty, k is not empty, and n is empty.

• The output is 2D with shape (m*rank_size, n), where rank_size indicates the number of devices.

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]: supports two, four, or eight devices and only the all mesh networking of HCCS links.

• [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]: supports two, four, eight, sixteen, or thirty-two devices and only the double ring networking of HCCS links.

• Ascend 950PR/Ascend 950DT:

  • Supports 2, 4, 8, 16, 32, and 64 devices and supports only the all-mesh networking of HCCS links.
  • The total data volume of AllGather(x1) collective communication cannot exceed 16 x 256 MB. The total data volume of collective communication is calculated as follows: m x k x sizeof(x1_dtype) x Number of devices. The internal implementation of the operator may vary according to the shape. Therefore, the total communication volume may be slightly less than this value.

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object]: The MC2 operator in a model supports only the same communicator.

[object Object]

The following example is for reference only. For details, see .

Note: This sample code calls some HCCL collective communication library APIs, including HcclGetCommName, HcclCommInitAll, and HcclCommDestroy. For details, see .

• [object Object]Atlas A2 training products/Atlas A2 inference products[object Object], [object Object]Atlas A3 training products/Atlas A3 inference products[object Object], and Ascend 950PR/Ascend 950DT:

  [object Object]