Re: Re:flink sql关联维表在lookup执行计划中的关联条件问题

[Jane Chan]

Hi iasiuide,

感谢提问. 先来回答最后一个问题

关联维表的限制条件有的会作为关联条件，有的不作为关联条件吗？ 这种有什么规律吗？
>

Lookup join 的 on condition 会在优化过程中经过一系列改写, 这里只简要对影响 lookup 和 where 的几处进行说明.

1. logical 阶段, FlinkFilterJoinRule 会将 on 条件 split 为针对单边的 (左表/右表) 和针对双边的.
**针对单边的 filter 会被尽量 pushdown 到 join 节点之前** (这意味着有可能会额外生成一个 Filter 节点);
Filter 节点后续如何变化取决于这个 filter 能否 pushdown 到 source, 如果不能, 那么在 physical
阶段它就会变成维表上面 Calc 节点 (denoted by calcOnTemporalTable) 里面的 condition.

2. 在 CommonPhysicalLookupJoin 里解析 allLookupKeys 的时候, 会试图从
calcOnTemporalTable 里把常量条件抽取出来形成最终的 lookup key (也就是 explain plan 里面
lookup=[...] 的内容), 在 explain 时, 只要存在 calcOnTemporalTable, where=[...]
就会被打印出来.

回到具体的 case

为什么关联第一张维表dim_ymfz_prod_sys_trans_log的限制条件AND b.trans_date = DATE_FORMAT
> (CURRENT_TIMESTAMP, 'MMdd') 在执行计划中，不作为 lookup的条件 ==>
> lookup=[bg_rel_trans_id=bg_rel_trans_id],
>

因为 b.trans_date = DATE_FORMAT (CURRENT_TIMESTAMP, 'MMdd')
是针对维表单边的条件且无法被下推. 另外, 这里使用了非确定性函数[1], 请关注结果的正确性.


> 关联第二张维表 dim_ptfz_ymfz_merchant_info 的限制条件ON b.member_id = c.pk_id AND
> c.data_source = 'merch' 在执行计划中，都是作为lookup的条件 ==>
> lookup=[data_source=_UTF-16LE'merch', pk_id=member_id],
>

此时常量可以被提取出来


> 关联第三张维表dim_ptfz_ymfz_merchant_info的限制条件 ON c.agent_id = d.pk_id AND
> (d.data_source = 'ex_agent' OR d.data_source = 'agent')
> 中关于data_source的条件，在执行计划中不是lookup的条件 ==> lookup=[pk_id=agent_id],
>

据我所知 lookup 目前应该还不支持 SARGable

[1]
https://nightlies.apache.org/flink/flink-docs-master/zh/docs/dev/table/concepts/determinism/

Best,
Jane

On Fri, Mar 8, 2024 at 11:19 AM iasiuide  wrote:

> 好的，已经贴了sql片段
>
> 在 2024-03-08 11:02:34，"Xuyang"  写道：
> >Hi, 你的图挂了，可以用图床或者直接贴SQL
> >
> >
> >
> >
> >--
> >
> >Best！
> >Xuyang
> >
> >
> >
> >
> >在 2024-03-08 10:54:19，"iasiuide"  写道：
> >
> >
> >
> >
> >
> >下面的sql片段中
> >ods_ymfz_prod_sys_divide_order  为kafka source表
> >dim_ymfz_prod_sys_trans_log   为mysql为表
> >dim_ptfz_ymfz_merchant_info   为mysql为表
> >
> >
> >
> >flink web ui界面的执行计划片段如下：
> >
> > [1]:TableSourceScan(table=[[default_catalog, default_database,
> ods_ymfz_prod_sys_divide_order, watermark=[-(CASE(IS NULL(create_time),
> 1970-01-01 00:00:00:TIMESTAMP(3), CAST(create_time AS TIMESTAMP(3))),
> 5000:INTERVAL SECOND)]]], fields=[row_kind, id, sys_date, bg_rel_trans_id,
> order_state, create_time, update_time, divide_fee_amt, divide_fee_flag])
> >+- [2]:Calc(select=[sys_date, bg_rel_trans_id, create_time,
> IF(SEARCH(row_kind, Sarg[_UTF-16LE'-D', _UTF-16LE'-U']), (-1 *
> divide_fee_amt), divide_fee_amt) AS div_fee_amt,
> Reinterpret(CASE(create_time IS NULL, 1970-01-01 00:00:00, CAST(create_time
> AS TIMESTAMP(3 AS ts], where=[((order_state = '2') AND (divide_fee_amt
>  0) AND (sys_date = DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS
> TIMESTAMP(9)), '-MM-dd')))])
> >   +-
> [3]:LookupJoin(table=[default_catalog.default_database.dim_ymfz_prod_sys_trans_log],
> joinType=[LeftOuterJoin], async=[false],
> lookup=[bg_rel_trans_id=bg_rel_trans_id], where=[(trans_date =
> DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS TIMESTAMP(9)), 'MMdd'))],
> select=[sys_date, bg_rel_trans_id, create_time, div_fee_amt, ts,
> bg_rel_trans_id, pay_type, member_id, mer_name])
> >  +- [4]:Calc(select=[sys_date, create_time, div_fee_amt, ts,
> pay_type, member_id, mer_name], where=[(CHAR_LENGTH(member_id)  1)])
> > +-
> [5]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
> joinType=[LeftOuterJoin], async=[false],
> lookup=[data_source=_UTF-16LE'merch', pk_id=member_id], where=[(data_source
> = 'merch')], select=[sys_date, create_time, div_fee_amt, ts, pay_type,
> member_id, mer_name, pk_id, agent_id, bagent_id])
> >+- [6]:Calc(select=[sys_date, create_time, div_fee_amt, ts,
> pay_type, member_id, mer_name, agent_id, bagent_id])
> >   +-
> [7]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
> joinType=[LeftOuterJoin], async=[false], lookup=[pk_id=agent_id],
> where=[SEARCH(data_source, Sarg[_UTF-16LE'agent', _UTF-16LE'ex_agent'])],
> select=[sys_date, create_time, div_fee_amt, ts, pay_type, member_id,
> mer_name, agent_id, bagent_id, pk_id, bagent_id, fagent_id])
> >  +- [8]:Calc(select=[sys_date, create_time, div_fee_amt,
> ts, pay_type, member_id, mer_name, bagent_id, bagent_id0, fagent_id AS
> fagent_id0])
> > +-
> [9]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
> joinType=[LeftOuterJoin], async=[false],
> lookup=[data_source=_UTF-16LE'agent', pk_id=fagent_id0],
> where=[(data_source = 'agent')], select=[sys_date, create_time,
> div_fee_amt, ts, pay_type, member_id, mer_name, bagent_id, bagent_id0,
> fagent_id0, pk_id, agent_name, bagent_name])
> >  
> >
> >
> >为什么关联第一张维表dim_ymfz_prod_sys_trans_log的限制条件AND b.trans_date = DATE_FORMAT
> (CURRENT_TIMESTAMP, 'MMdd') 在执行计划中，不作为 lookup的条件 ==>
> lookup=[bg_rel_trans_id=bg_rel_trans_id],
> >关联第二张维表 dim_ptfz_ymfz_merchant_info 的限制条件ON b.member_id = c.pk_id AND
> 

Re:flink sql关联维表在lookup执行计划中的关联条件问题

[Xuyang]

Hi, 你的图挂了，可以用图床或者直接贴SQL




--

Best！
Xuyang




在 2024-03-08 10:54:19，"iasiuide"  写道：





下面的sql片段中
ods_ymfz_prod_sys_divide_order  为kafka source表
dim_ymfz_prod_sys_trans_log   为mysql为表
dim_ptfz_ymfz_merchant_info   为mysql为表



flink web ui界面的执行计划片段如下：

 [1]:TableSourceScan(table=[[default_catalog, default_database, 
ods_ymfz_prod_sys_divide_order, watermark=[-(CASE(IS NULL(create_time), 
1970-01-01 00:00:00:TIMESTAMP(3), CAST(create_time AS TIMESTAMP(3))), 
5000:INTERVAL SECOND)]]], fields=[row_kind, id, sys_date, bg_rel_trans_id, 
order_state, create_time, update_time, divide_fee_amt, divide_fee_flag])
+- [2]:Calc(select=[sys_date, bg_rel_trans_id, create_time, IF(SEARCH(row_kind, 
Sarg[_UTF-16LE'-D', _UTF-16LE'-U']), (-1 * divide_fee_amt), divide_fee_amt) AS 
div_fee_amt, Reinterpret(CASE(create_time IS NULL, 1970-01-01 00:00:00, 
CAST(create_time AS TIMESTAMP(3 AS ts], where=[((order_state = '2') AND 
(divide_fee_amt  0) AND (sys_date = DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS 
TIMESTAMP(9)), '-MM-dd')))])
   +- 
[3]:LookupJoin(table=[default_catalog.default_database.dim_ymfz_prod_sys_trans_log],
 joinType=[LeftOuterJoin], async=[false], 
lookup=[bg_rel_trans_id=bg_rel_trans_id], where=[(trans_date = 
DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS TIMESTAMP(9)), 'MMdd'))], 
select=[sys_date, bg_rel_trans_id, create_time, div_fee_amt, ts, 
bg_rel_trans_id, pay_type, member_id, mer_name])
  +- [4]:Calc(select=[sys_date, create_time, div_fee_amt, ts, pay_type, 
member_id, mer_name], where=[(CHAR_LENGTH(member_id)  1)])
 +- 
[5]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[data_source=_UTF-16LE'merch', 
pk_id=member_id], where=[(data_source = 'merch')], select=[sys_date, 
create_time, div_fee_amt, ts, pay_type, member_id, mer_name, pk_id, agent_id, 
bagent_id])
+- [6]:Calc(select=[sys_date, create_time, div_fee_amt, ts, 
pay_type, member_id, mer_name, agent_id, bagent_id])
   +- 
[7]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[pk_id=agent_id], 
where=[SEARCH(data_source, Sarg[_UTF-16LE'agent', _UTF-16LE'ex_agent'])], 
select=[sys_date, create_time, div_fee_amt, ts, pay_type, member_id, mer_name, 
agent_id, bagent_id, pk_id, bagent_id, fagent_id])
  +- [8]:Calc(select=[sys_date, create_time, div_fee_amt, ts, 
pay_type, member_id, mer_name, bagent_id, bagent_id0, fagent_id AS fagent_id0])
 +- 
[9]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[data_source=_UTF-16LE'agent', 
pk_id=fagent_id0], where=[(data_source = 'agent')], select=[sys_date, 
create_time, div_fee_amt, ts, pay_type, member_id, mer_name, bagent_id, 
bagent_id0, fagent_id0, pk_id, agent_name, bagent_name])
  


为什么关联第一张维表dim_ymfz_prod_sys_trans_log的限制条件AND b.trans_date = DATE_FORMAT 
(CURRENT_TIMESTAMP, 'MMdd') 在执行计划中，不作为 lookup的条件 ==> 
lookup=[bg_rel_trans_id=bg_rel_trans_id],
关联第二张维表 dim_ptfz_ymfz_merchant_info 的限制条件ON b.member_id = c.pk_id AND 
c.data_source = 'merch' 在执行计划中，都是作为lookup的条件 ==> 
lookup=[data_source=_UTF-16LE'merch', pk_id=member_id],
关联第三张维表dim_ptfz_ymfz_merchant_info的限制条件 ON c.agent_id = d.pk_id AND 
(d.data_source = 'ex_agent' OR d.data_source = 'agent') 
中关于data_source的条件，在执行计划中不是lookup的条件 ==> lookup=[pk_id=agent_id],
关联维表的限制条件有的会作为关联条件，有的不作为关联条件吗？ 这种有什么规律吗？ 因为这个会关乎维表的索引字段的设置。

Re:flink sql关联维表在lookup执行计划中的关联条件问题

[iasiuide]

图片可能加载不出来，下面是图片中的sql片段 
 ..
 END AS trans_type,

  a.div_fee_amt,

  a.ts

FROM

  ods_ymfz_prod_sys_divide_order a

  LEFT JOIN dim_ymfz_prod_sys_trans_log FOR SYSTEM_TIME AS OF a.proc_time 
AS b ON a.bg_rel_trans_id = b.bg_rel_trans_id

  AND b.trans_date = DATE_FORMAT (CURRENT_TIMESTAMP, 'MMdd')

  LEFT JOIN dim_ptfz_ymfz_merchant_info FOR SYSTEM_TIME AS OF a.proc_time 
AS c ON b.member_id = c.pk_id

  AND c.data_source = 'merch'

  LEFT JOIN dim_ptfz_ymfz_merchant_info FOR SYSTEM_TIME AS OF a.proc_time 
AS d ON c.agent_id = d.pk_id

  AND (

d.data_source = 'ex_agent'

OR d.data_source = 'agent'

  ) 

  LEFT JOIN dim_ptfz_ymfz_merchant_info FOR SYSTEM_TIME AS OF a.proc_time 
AS d1 ON d.fagent_id = d1.pk_id

  AND d1.data_source = 'agent'

WHERE 

  a.order_state = '2' 

  AND a.divide_fee_amt > 0

  ) dat

WHERE

  trans_date = DATE_FORMAT (CURRENT_TIMESTAMP, '-MM-dd')

  AND CHAR_LENGTH(member_id) > 1;













在 2024-03-08 10:54:19，"iasiuide"  写道：





下面的sql片段中
ods_ymfz_prod_sys_divide_order  为kafka source表
dim_ymfz_prod_sys_trans_log   为mysql为表
dim_ptfz_ymfz_merchant_info   为mysql为表



flink web ui界面的执行计划片段如下：

 [1]:TableSourceScan(table=[[default_catalog, default_database, 
ods_ymfz_prod_sys_divide_order, watermark=[-(CASE(IS NULL(create_time), 
1970-01-01 00:00:00:TIMESTAMP(3), CAST(create_time AS TIMESTAMP(3))), 
5000:INTERVAL SECOND)]]], fields=[row_kind, id, sys_date, bg_rel_trans_id, 
order_state, create_time, update_time, divide_fee_amt, divide_fee_flag])
+- [2]:Calc(select=[sys_date, bg_rel_trans_id, create_time, IF(SEARCH(row_kind, 
Sarg[_UTF-16LE'-D', _UTF-16LE'-U']), (-1 * divide_fee_amt), divide_fee_amt) AS 
div_fee_amt, Reinterpret(CASE(create_time IS NULL, 1970-01-01 00:00:00, 
CAST(create_time AS TIMESTAMP(3 AS ts], where=[((order_state = '2') AND 
(divide_fee_amt  0) AND (sys_date = DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS 
TIMESTAMP(9)), '-MM-dd')))])
   +- 
[3]:LookupJoin(table=[default_catalog.default_database.dim_ymfz_prod_sys_trans_log],
 joinType=[LeftOuterJoin], async=[false], 
lookup=[bg_rel_trans_id=bg_rel_trans_id], where=[(trans_date = 
DATE_FORMAT(CAST(CURRENT_TIMESTAMP() AS TIMESTAMP(9)), 'MMdd'))], 
select=[sys_date, bg_rel_trans_id, create_time, div_fee_amt, ts, 
bg_rel_trans_id, pay_type, member_id, mer_name])
  +- [4]:Calc(select=[sys_date, create_time, div_fee_amt, ts, pay_type, 
member_id, mer_name], where=[(CHAR_LENGTH(member_id)  1)])
 +- 
[5]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[data_source=_UTF-16LE'merch', 
pk_id=member_id], where=[(data_source = 'merch')], select=[sys_date, 
create_time, div_fee_amt, ts, pay_type, member_id, mer_name, pk_id, agent_id, 
bagent_id])
+- [6]:Calc(select=[sys_date, create_time, div_fee_amt, ts, 
pay_type, member_id, mer_name, agent_id, bagent_id])
   +- 
[7]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[pk_id=agent_id], 
where=[SEARCH(data_source, Sarg[_UTF-16LE'agent', _UTF-16LE'ex_agent'])], 
select=[sys_date, create_time, div_fee_amt, ts, pay_type, member_id, mer_name, 
agent_id, bagent_id, pk_id, bagent_id, fagent_id])
  +- [8]:Calc(select=[sys_date, create_time, div_fee_amt, ts, 
pay_type, member_id, mer_name, bagent_id, bagent_id0, fagent_id AS fagent_id0])
 +- 
[9]:LookupJoin(table=[default_catalog.default_database.dim_ptfz_ymfz_merchant_info],
 joinType=[LeftOuterJoin], async=[false], lookup=[data_source=_UTF-16LE'agent', 
pk_id=fagent_id0], where=[(data_source = 'agent')], select=[sys_date, 
create_time, div_fee_amt, ts, pay_type, member_id, mer_name, bagent_id, 
bagent_id0, fagent_id0, pk_id, agent_name, bagent_name])
  


为什么关联第一张维表dim_ymfz_prod_sys_trans_log的限制条件AND b.trans_date = DATE_FORMAT 
(CURRENT_TIMESTAMP, 'MMdd') 在执行计划中，不作为 lookup的条件 ==> 
lookup=[bg_rel_trans_id=bg_rel_trans_id],
关联第二张维表 dim_ptfz_ymfz_merchant_info 的限制条件ON b.member_id = c.pk_id AND 
c.data_source = 'merch' 在执行计划中，都是作为lookup的条件 ==> 
lookup=[data_source=_UTF-16LE'merch', pk_id=member_id],
关联第三张维表dim_ptfz_ymfz_merchant_info的限制条件 ON c.agent_id = d.pk_id AND 
(d.data_source = 'ex_agent' OR d.data_source = 'agent') 
中关于data_source的条件，在执行计划中不是lookup的条件 ==> lookup=[pk_id=agent_id],
关联维表的限制条件有的会作为关联条件，有的不作为关联条件吗？ 这种有什么规律吗？ 因为这个会关乎维表的索引字段的设置。