Using git bisect it was found that the regression was caused by the initial commits of MCOL-523

An attempt to optimize the code was bad logic.

So what was done to this ticket?

Part of MCOL-523 was reverted (an attempt at a performance improvement). The test for this is TPCDS Query 17.

Build verified: 1.1.1-1 GitHub source

/root/columnstore/mariadb-columnstore-server
commit 3d846d3277ba970b32ba3f471323fcac58b5c35d
Author: david hill <david.hill@mariadb.com>
Date: Mon Oct 23 09:57:05 2017 -0500

change to 1.1.1

/root/columnstore/mariadb-columnstore-server/mariadb-columnstore-engine
commit 4aa7eb1830ddf585706f804b1982589c7d67ff0a
Author: root <root@srvhill01.lan>
Date: Mon Oct 23 09:56:07 2017 -0500

change to 1.1.1

Build the tpcds database and loaded 1gb dataset.

MariaDB [tpcds]> select i_item_id
-> ,i_item_desc
-> ,s_state
-> ,count(ss_quantity) as store_sales_quantitycount
-> ,avg(ss_quantity) as store_sales_quantityave
-> ,stddev_samp(ss_quantity) as store_sales_quantitystdev
-> ,stddev_samp(ss_quantity)/avg(ss_quantity) as store_sales_quantitycov
-> ,count(sr_return_quantity) as store_returns_quantitycount
-> ,avg(sr_return_quantity) as store_returns_quantityave
-> ,stddev_samp(sr_return_quantity) as store_returns_quantitystdev
-> ,stddev_samp(sr_return_quantity)/avg(sr_return_quantity) as store_returns_quantitycov
-> ,count(cs_quantity) as catalog_sales_quantitycount ,avg(cs_quantity) as catalog_sales_quantityave
-> ,stddev_samp(cs_quantity) as catalog_sales_quantitystdev
-> ,stddev_samp(cs_quantity)/avg(cs_quantity) as catalog_sales_quantitycov
-> from store_sales
-> ,store_returns
-> ,catalog_sales
-> ,date_dim d1
-> ,date_dim d2
-> ,date_dim d3
-> ,store
-> ,item
-> where d1.d_quarter_name = '1998Q1'
-> and d1.d_date_sk = ss_sold_date_sk
-> and i_item_sk = ss_item_sk
-> and s_store_sk = ss_store_sk
-> and ss_customer_sk = sr_customer_sk
-> and ss_item_sk = sr_item_sk
-> and ss_ticket_number = sr_ticket_number
-> and sr_returned_date_sk = d2.d_date_sk
-> and d2.d_quarter_name in ('1998Q1','1998Q2','1998Q3')
-> and sr_customer_sk = cs_bill_customer_sk
-> and sr_item_sk = cs_item_sk
-> and cs_sold_date_sk = d3.d_date_sk
-> and d3.d_quarter_name in ('1998Q1','1998Q2','1998Q3')
-> group by i_item_id
-> ,i_item_desc
-> ,s_state
-> order by i_item_id
-> ,i_item_desc
-> ,s_state
-> limit 100;
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1 row in set (0.95 sec)