buf_pool_cfg

Applicability

Product

Supported (Yes/No)

Atlas 350 Accelerator Card

No

Atlas A3 training product/Atlas A3 inference product

Yes

Atlas A2 training product/Atlas A2 inference product

Yes

Atlas 200I/500 A2 inference product

No

Atlas inference product

No

Atlas training product

No

Note: For the Atlas A2 training product/Atlas A2 inference product, only the Atlas 800I A2 inference server and A200I A2 Box heterogeneous subrack are supported.

Function Description

Specifies the memory level configuration to improve memory allocation performance and utilization.

Prototype

1
buf_pool_cfg(buf_pool_cfg)

Parameters

Parameter

Data Type

Description

buf_pool_cfg

string

Character string in JSON array format, including the total_size, blk_size, and max_buf_size nodes.

Configuration Item

Optional/Mandatory

Description

total_size

Mandatory

Size of the memory pool at the current level, in bytes.

Constraints:

The value of total_size is a multiple of 2 MB, and the value of total_size is a multiple of blk_size. The maximum value cannot exceed the value of 0xFFFFFFFF.

blk_size

Mandatory

Minimum memory that can be allocated at a time at the current level, in bytes.

Constraints:

The value must be 2n within a range of (0, 2 MB], and less than or equal to the value of max_buf_size.

max_buf_size

Mandatory

Maximum memory that can be allocated at a time at the current level, in bytes.

Constraints: The value must be less than that of total_size. If multiple levels are configured, the max_buf_size values must be strictly increasing across levels.

Example

1
2
3
from llm_datadist import LLMConfig
llm_config = LLMConfig()
llm_config.buf_pool_cfg= '{"buf_cfg": [{"total_size":2097152,"blk_size":256,"max_buf_size":8192},{"total_size": 10485760,"blk_size": 8192,"max_buf_size": 8388608},{"total_size": 69206016,"blk_size": 8192,"max_buf_size": 67108864}]}'

Returns

In normal cases, no value is returned.

If a parameter is incorrect, a TypeError or ValueError may be thrown.

Constraints

None