Doris主键点查比Clickhouse慢一个数量级原因

Doris从一个400亿行的表根据主键查询, 耗时2秒

mysql> select count(1) from b_YGF5JvJfvkccIVUai68nk0;
+-------------+
| count(1)    |
+-------------+
| 45791646046 |
+-------------+
1 row in set (0.94 sec)

mysql> select * from b_YGF5JvJfvkccIVUai68nk0  where token = 'infozhanganfahuadandanzhangkaiyan';

Empty set (2.00 sec)

Clickhouse从一个4000亿行的表根据主键查询, 耗时0.2秒

select count(1) from b_P9PCCRQsytjxP4P9zBI7od;

SELECT count(1)
FROM b_P9PCCRQsytjxP4P9zBI7od

Query id: 2cb19c19-4cec-46de-bfdc-5608ad1c528b

┌──────count()─┐
│ 409913993810 │
└──────────────┘

1 row in set. Elapsed: 0.008 sec. 

bjdd-bls-ck-online-017.bjdd :) 
bjdd-bls-ck-online-017.bjdd :) select * from b_P9PCCRQsytjxP4P9zBI7od  where token = 'infozhanganfahuadandanzhangkaiyan';

SELECT *
FROM b_P9PCCRQsytjxP4P9zBI7od
WHERE token = 'infozhanganfahuadandanzhangkaiyan'

Query id: 562601d0-ef96-43ab-8019-431bae37229f

Ok.

0 rows in set. Elapsed: 0.201 sec. Processed 126.05 thousand rows, 2.81 MB (625.61 thousand rows/s., 13.94 MB/s.)

都是前缀索引的第一位查询 (第一位是高基数类似于request_id), 而且给Doris灌的数据还少很多, 为什么性能比clickhouse差这么多呢