Details
Description
Steps to reproduce -
- Create a 2 node cluster with 1 KV+N1QL and 1 CBAS node.
- Create buckets to load TPCH data using cr_bucket.sh
- Load data into buckets using import.sh
- Create analytics dataset on all the buckets created in step 2 using cr_dataset.sql
- Now create samples on all the dataset with sample-seed value set to 10 (using this seed value for my automation)
- Now run following query on analytics workbench and check the query plan
SET `compiler.queryplanshape` "rightdeep";
|
select s.s_acctbal,
|
s.s_name,
|
n.n_name,
|
p.p_partkey,
|
p.p_mfgr,
|
s.s_address,
|
s.s_phone,
|
s.s_comment
|
from part p,
|
supplier s,
|
partsupp ps,
|
nation n,
|
region r
|
where p.p_partkey = ps.ps_partkey
|
and s.s_suppkey = ps.ps_suppkey
|
and p.p_size = 37
|
and p.p_type like '%COPPER'
|
and s.s_nationkey = n.n_nationkey
|
and n.n_regionkey = r.r_regionkey
|
and r.r_name = 'EUROPE'
|
and ps.ps_supplycost = (select VALUE min(ps2.ps_supplycost)
|
from part p2,
|
partsupp ps2,
|
supplier s2,
|
nation n2,
|
region r2
|
where p2.p_partkey = ps2.ps_partkey
|
and p2.p_partkey = p.p_partkey
|
and s2.s_suppkey = ps2.ps_suppkey
|
and s2.s_nationkey = n2.n_nationkey
|
and n2.n_regionkey = r2.r_regionkey
|
and r2.r_name = 'EUROPE')[0] order by s.s_acctbal desc,
|
n.n_name,
|
s.s_name,
|
p.p_partkey limit 100;
|
As can be seen in the screen recording, the Query plan window is not utilised completely to display the query plan when zoomed in.