query_register_mem_status
函数功能
在开启CacheManager的场景下,开启方式是在初始化LLMDataDist时传入llm.EnableCacheManager为1时,调用此接口查询注册内存状态。
函数原型
query_register_mem_status(comm_id: int) -> RegisterMemStatus
参数说明
参数名称 |
数据类型 |
取值说明 |
|---|---|---|
comm_id |
int |
link接口的返回值 |
调用示例
from llm_datadist import LLMDataDist, LLMRole, LlmConfig
device_id = 0
llm_datadist = LLMDataDist(LLMRole.PROMPT, 1)
options = {
"ge.exec.deviceId": str(device_id),
"llm.MemPoolConfig": "{\"memory_size\": 1073741824}"
}
llm_config = LlmConfig()
llm_config.enable_cache_manager = True
options.update(llm_config.generate_options())
llm_datadist.init(options)
rank_table = """
{
"server_count": "2",
"server_list": [
{
"device": [
{
"device_id": "0",
"device_ip": "x.x.x.x",
"rank_id": "0"
},
],
"server_id": "x.x.x.x"
},{
"device": [
{
"device_id": "0",
"device_ip": "x.x.x.x",
"rank_id": "0"
},
],
"server_id": "x.x.x.x"
}
],
"status": "completed",
"version": "1.0"
}"""
comm_id = llm_datadist.link({1: 0, 2: 1}, rank_table)
print("init link suc:", comm_id)
while True:
ret = llm_datadist.query_register_mem_status(comm_id)
if ret == RegisterMemStatus.OK:
print("query_register_mem_status ok", flush=True)
break
elif ret == RegisterMemStatus.FAILED:
print("query_register_mem_status failed", flush=True)
raise RuntimeError("link failed")
print("need check again", flush=True)
time.sleep(1)
返回值
无。
异常场景会抛出LLMException异常。
约束说明
需要开启CacheManager模式。
父主题: LLMDataDist