DataWorks找阿里运维截图提示如图所示，看描述是cpu资源沾满了么？

DataWorks现在的pyodps脚本固化历史分区数据执行超过48小时执行不完，看日志，只跑了1/5的样子，找阿里运维截图提示如图所示，看描述是cpu资源沾满了么？我们数据量并不大，8000个分区，2.7亿条数据而已
pyodps大致操作如下
import datetime
from odps import ODPS
import pandas as pd

sql =
SELECT DISTINCT pt
FROM p.XXXXXXXta.st.XXXXXXXtatai order by pt desc

instance = odps.run_sql(sql)
instance.wait_for_success()

results = []

with instance.open_reader() as reader:
for record in reader:
results.append(record.values)

distinct_pt_df = pd.DataFrame(results, columns=['pt'])

total_partitions = len(distinct_pt_df)

for index, row in distinct_pt_df.iterrows():
partition_str = str(row['pt'])

print(f"当前执行序号：{index+1} / 总个数：{total_partitions}")
sql_insert = f
INSERT OVERWRITE TABLE sc_dr_ytai_sjzq.STG_SD_YT_SETL_FEE_STT_D PARTITION(pt='{partition_str}')
select XXXXXXXta
from scXXXXXXXtaqXXXXXXXtaai c
where XXXXXXXtaloc not in ('3', '4')
and a.psn_no = d.psn_no;

print (sql_insert )
insert_instance = odps.run_sql(sql_insert )
insert_instance.wait_for_success()!