I use SQLServer 2014 (120 compatibility mode) and I want a parallel execution plan for my query, but because of a few fields, it isn't. As long as field pi.Name AS ProcessName is commented I have parallelism but when it is active it isn't. The same problem is with other commented fields. Is the reason connected with an index which covers pi.ProcessGuid but not pi.Name or something more?
ExecutionPlan_Parallel
ExecutionPlan_Single
ExecutionPlan_Original
When I commented on those few fields ExecutionPlan starts to be parallel and from 30 minutes query runs only 11 seconds.
SELECT kg.Id,
ui.UserInfoGuid AS UserGuid,
ui.FullName AS UserName,
di.DeviceInfoGuid AS DeviceGuid,
di.ComputerFullName,
pi.ProcessGuid,
--pi.Name AS ProcessName,
kg.ProcessActivationGuid,
t.Caption,
whi.WebsiteHostGuid,
--whi.HostName,
wv.WebsiteGuid,
wv.[Url],
--cc.CategoryGuid,
--cc.[Name] AS CategoryName,
ca.ControlSequenceGuid,
ca.ControlActivationGuid/*
(SELECT ce.[Name] + '/' AS [text()]
FROM [raw].ControlElement AS ce
WHERE ce.ControlSequenceGuid = ca.ControlSequenceGuid
ORDER BY ce.SequenceNumber DESC
FOR xml path ('')) AS ControlTree*/
--,ac.ApplicationContextName
,kg.BeginDate AS KeystrokeBeginDate
,kg.EndDate AS KeystrokeEndDate
,kg.TotalKeyCount
,kg.KeyCount
,kg.FunctionalKeyCount
,kg.ClickCount
,kg.WheelCount
,kg.OtherKeyCount
,kg.TotalMilliseconds
,kg.ActiveMilliseconds
FROM raw.[Session] AS s
INNER JOIN raw.Monitoring AS m
ON ( m.SessionGuid = s.SessionGuid )
INNER JOIN stats.[KeystrokeGroupRcpView] AS kg
ON ( kg.MonitoringGuid = m.MonitoringGuid )
INNER JOIN dbo.UserInfo AS ui
ON ( ui.UserInfoGuid = kg.UserGuid )
INNER JOIN dbo.DeviceInfo AS di
ON ( di.DeviceInfoGuid = kg.DeviceGuid )
INNER JOIN dbo.ProcessInfo AS pi
ON ( pi.OrganizationGuid = di.OrganizationGuid
AND pi.ProcessGuid = kg.ProcessGuid )
INNER JOIN raw.Title AS t
ON ( t.TitleGuid = kg.TitleGuid )
LEFT OUTER JOIN [raw].ControlActivation AS ca
ON ca.ProcessActivationGuid = kg.ProcessActivationGuid
AND kg.BeginDate >= ca.BeginDate
AND kg.EndDate <= ca.EndDate
LEFT OUTER JOIN controls.ControlFocus AS cf
ON cf.ControlActivationGuid = ca.ControlActivationGuid
LEFT OUTER JOIN controls.ControlSequenceContext AS csc
ON csc.ControlSequenceGuid = ca.ControlSequenceGuid
LEFT OUTER JOIN controls.ApplicationContext AS ac
ON ac.ApplicationContextGuid = csc.ApplicationContextGuid
LEFT OUTER JOIN controls.ProcessControlCategory AS pcc
ON pcc.ProcessGuid = pi.ProcessGuid
AND pcc.ControlCategoryKeyGuid =
cf.ControlCategoryKeyGuid
LEFT OUTER JOIN controls.ControlCategory AS cc
ON cc.CategoryGuid = pcc.CategoryGuid
LEFT OUTER JOIN raw.WebsiteVisit AS wv
ON ( pi.IsWebBrowser = 1
AND wv.ProcessActivationGuid =
kg.ProcessActivationGuid
AND ( wv.BeginDate < kg.BeginDate
AND wv.EndDate >= kg.EndDate ) )
LEFT OUTER JOIN dbo.WebsiteHostInfo AS whi
ON ( whi.WebsiteHostGuid = wv.WebsiteHostGuid
AND whi.OrganizationGuid = di.OrganizationGuid )
WHERE kg.BeginDate > '2020-07-01' AND kg.BeginDate < '2020-07-02' and kg.UserGuid in ('A170565A-2D30-4911-5B9C-8525E2A2772B','4BE982BD-ADFC-6201-31C2-60A0BEF0D7F6','AE296576-87C6-EC2F-5A6F-E24ACE14456E')
Related
My select statement returns the ids from the tables I have joined correctly(393, starting at 92 and going up 484, its not a PK field, so some ids can be used more then once) there is a total of 550 rows I need to update, when I run the update it only adds the number 92 to the table for all 550 rows.
update movements set [location] = locaID.id FROM (
SELECT
lo.id
FROM [Harvest_Transactions] ht
left join [Harvest_Master_Shift] hms on ht.Harvest_Master_id = hms.Harvest_Master_id
left join [Greenhouse_Troughs] gt on ht.Trough = gt.Greenhouse_Trough_id
left join [batches] b on b.name = hms.Batch_id
left join [phases] p on p.batch = b.id and p.[type] = 3
left join #loc lo on lo.name = gt.Trough_No and lo.area = gt.SQM_per_Trough and lo.Bay_id = gt.Bay_id
where ht.Number_of_Plants_Harvested != 0 and (hms.CreatedOn <= '2020-02-05 09:33:00.000' OR hms.CreatedOn is null )
)locaID where product = 14
what am I missing so it updates with the correct values?
My first impression of your query is that only the rows with product = 14 in movements table will be updated with the single value coming from the subquery.
If you want to update the values based on another derived table, you need some way to link the rows together - i.e a JOIN between your table being updated and the table that holds the other data (see here )
What is the common column between movements and
SELECT
lo.id
FROM [Harvest_Transactions] ht
left join [Harvest_Master_Shift] hms on ht.Harvest_Master_id = hms.Harvest_Master_id
left join [Greenhouse_Troughs] gt on ht.Trough = gt.Greenhouse_Trough_id
left join [batches] b on b.name = hms.Batch_id
left join [phases] p on p.batch = b.id and p.[type] = 3
left join #loc lo on lo.name = gt.Trough_No and lo.area = gt.SQM_per_Trough and
lo.Bay_id = gt.Bay_id
where ht.Number_of_Plants_Harvested != 0 and (hms.CreatedOn <= '2020-02-05
09:33:00.000' OR hms.CreatedOn is null )
The rough syntax shouldI think be
UPDATE movements
set [location] = locaID.id
FROM movements
JOIN (
SELECT
lo.id
FROM [Harvest_Transactions] ht
left join [Harvest_Master_Shift] hms on ht.Harvest_Master_id = hms.Harvest_Master_id
left join [Greenhouse_Troughs] gt on ht.Trough = gt.Greenhouse_Trough_id
left join [batches] b on b.name = hms.Batch_id
left join [phases] p on p.batch = b.id and p.[type] = 3
left join #loc lo on lo.name = gt.Trough_No and lo.area = gt.SQM_per_Trough and
lo.Bay_id = gt.Bay_id
where ht.Number_of_Plants_Harvested != 0 and (hms.CreatedOn <= '2020-02-05
09:33:00.000' OR hms.CreatedOn is null )
) locaID
ON movements.<some column> = locaID.<some column>
Is correct to put these conditional sentences in join statement in order to improve performance?
I just tried to use same code using one union and cross apply per condition used, but when I use the code in other queries, is too slow.
SELECT u2.fkPropertyID, u2.pkUnitId, u2.UnitCode,
u2.fkUnitViewID,u2.fkFloorPlanID, u2.fkCalendarTypeID
FROM dbo.tblUnit u
left join dbo.tblUnitGroup ug ON ug.fkUnitLogicalID = u.pkUnitID
left JOIN dbo.tblUnit u2 on ((u2.fkUnitGroupID = u.pkUnitID and
u.ynPhyUnit = 0)
or
(u2.pkUnitID = u.pkUnitID and u2.ynPhyUnit =1))
or
(u2.pkUnitId = ug.fkUnitPhisicallID and u2.ynPhyUnit = 1)
where u.pkUnitId = 3806
Just for fun here is one way you could rewrite this. I suspect if you look at the execution plan it will be identical.
SELECT u2.fkPropertyID
, u2.pkUnitId
, u2.UnitCode
, u2.fkUnitViewID
, u2.fkFloorPlanID
, u2.fkCalendarTypeID
FROM dbo.tblUnit u
left join dbo.tblUnitGroup ug ON ug.fkUnitLogicalID = u.pkUnitID
left JOIN dbo.tblUnit u2 on
(
u2.fkUnitGroupID = u.pkUnitID
and
u.ynPhyUnit = 0
)
or
(
u2.pkUnitID in (u.pkUnitID, ug.fkUnitPhisicallID)
and
u2.ynPhyUnit = 1
)
where u.pkUnitId = 3806
I get this with best results in execution plan, I am reviewing with other values:
SELECT u2.fkPropertyID, u2.pkUnitId, u2.UnitCode, u2.fkUnitViewID,
u2.fkFloorPlanID, u2.fkCalendarTypeID
FROM tblUnit u
left join dbo.tblUnitGroup ug ON ug.fkUnitLogicalID = u.pkUnitID
outer apply
(
select uu2.fkPropertyID, uu2.pkUnitId, uu2.UnitCode,
uu2.fkUnitViewID, uu2.fkFloorPlanID, uu2.fkCalendarTypeID
from dbo.tblUnit uu2
where ( ((uu2.fkUnitGroupID = u.pkUnitID) and (u.ynPhyUnit = 0))
or
((uu2.pkUnitID = u.pkUnitID) and (uu2.ynPhyUnit = 1)) )
or
( (uu2.pkUnitId = ug.fkUnitPhisicallID) and (uu2.ynPhyUnit = 1) )
) as u2
where u.pkUnitId = 3806
I have a sql that is as under:
SELECT ib.branch_no,
ib.on_hand,
p.weightedav,
p.item_code,
FROM physical p
INNER JOIN
item_branch as ib on p.item_code = ib.item_code
WHERE ib.on_hand <> 0
This SQL returns only those branch_no that have on_hand <> 0.
I am trying to get all the branch_nos irrespective of the on_hand field, but while still using the where on_hand clause.
Taking the on_hand clause away solves my problem, but gives me large amount of un-needed rows with 0's.
I am using SQL SERVER 2008 R2.
Thanks in advance for any guidance. Please apologize if I am missing any information.
------------------------------------------ENTIRE SQL QUERY (Updated)---------------------------------------------
select
ib.branch_no,
p.weighted_av,
p.item_code,
p.x_value,
p.y_value,
ib.on_hand,
p.on_hand as PhysicalOH,
ip.price,
i.item_code as StyleCode,
i.description,
i.cat1,
i.cat2,
i.cat3,
i.cat4,
np.is_style_yn,
si.supplier_code ,
ysv.sort as YSort
from physical as p
left outer JOIN
item_branch as ib on p.item_code = ib.item_code -- and ib.on_hand <> 0
INNER JOIN
item_price as ip on p.item_code = ip.item_code and ip.price_type = 'P1'
INNER JOIN
style_values as sv on p.style_code = sv.style_code and p.x_value = sv.value
INNER JOIN
style_values as ysv on p.style_code = ysv.style_code and p.y_value = ysv.value and ysv.axis = 'Y'
INNER JOIN
ITEM as i on p.style_code = i.item_code
INNER JOIN
NON_PHYSICAL as np ON i.item_code = np.item_code and np.is_style_yn = 1
INNER JOIN
supplier_item as si ON i.item_code = si.item_code and si.pref_supp_no = 1
where --ib.on_hand <> 0 and
sv.axis = 'X' and
i.item_code in
(SELECT ITEM.item_code
FROM ITEM
INNER JOIN
NON_PHYSICAL ON ITEM.item_code = NON_PHYSICAL.item_code
LEFT JOIN
supplier_item ON Item.item_code = supplier_item.item_code and pref_supp_no = 1
WHERE NON_PHYSICAL.is_style_yn = 1 and ITEM.cat1 = 'Verge Sportswear Ltd' )
order by
si.supplier_code,
i.cat4,
i.cat3,
i.cat2,
i.cat1,
sv.sort
Try this:
SELECT ib.branch_no,
ib.on_hand,
p.weightedav,
p.item_code,
FROM physical p
INNER JOIN
item_branch as ib on (p.item_code = ib.item_code AND ib.on_hand <> 0)
I had a view that worked, I edited it in SQL Server Mgmt Studio, it didn't work, I took out the changes and saved it down, it still doesn't work despite reverting it back to its original state.
I'm getting the error
Joined tables cannot be specified in a query containing outer join
operators. View or function 'adept_invoiced' contains joined tables.
I've googled around and come up with a possible join syntax problem due to the view being written in 2001 and me trying to edit it (for the first time) in Management Studio.
The view ran perfectly for years until I edited it this morning.
This is the From/Where section....
FROM dbo.SYS_COMP_ADDRESS
CROSS JOIN dbo.SL_ACCOUNTS
INNER JOIN dbo.SL_TRANSACTIONS
INNER JOIN dbo.ORD_HEADER
ON dbo.SL_TRANSACTIONS.ST_ORDER_NUMBER =
dbo.ORD_HEADER.OH_ORDER_NUMBER
INNER JOIN dbo.SL_PL_NL_DETAIL
LEFT OUTER JOIN dbo.STK_STOCK
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK.STKCODE
INNER JOIN dbo.ORD_DETAIL
ON dbo.SL_PL_NL_DETAIL.DET_ORDER_LINK =
dbo.ORD_DETAIL.OD_PRIMARY
LEFT OUTER JOIN dbo.STK_STOCK3
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK3.STKCODE3
LEFT OUTER JOIN dbo.STK_STOCK_2
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK_2.STKCODE2
LEFT OUTER JOIN dbo.PRC_PRICE_RECS2
ON dbo.SL_PL_NL_DETAIL.DET_PRICE_CODE =
dbo.PRC_PRICE_RECS2.PRCODE2
ON dbo.SL_TRANSACTIONS.ST_HEADER_KEY =
dbo.SL_PL_NL_DETAIL.DET_HEADER_KEY
ON dbo.SL_ACCOUNTS.CUCODE = dbo.SL_TRANSACTIONS.ST_COPYCUST
INNER JOIN dbo.SL_ACCOUNTS2
ON dbo.SL_ACCOUNTS.CUCODE = dbo.SL_ACCOUNTS2.CUCODE2
WHERE ( dbo.SL_TRANSACTIONS.ST_TRANTYPE IN ( 'INV', 'CRN' ) )
AND ( dbo.SL_TRANSACTIONS.ST_BATCH_FLAG <> 1 )
AND ( dbo.SL_ACCOUNTS.CUCODE <> '1023134' )
AND ( dbo.SL_ACCOUNTS.CUCODE <> '1023265' )
AND ( dbo.SL_ACCOUNTS.CUSORT NOT LIKE '%other income%' )
AND ( dbo.SL_ACCOUNTS.CUSORT NOT LIKE '%supplier%' )
I'm not sure where it might be failing, or how to fix it!
Any help would be greatly appreciated.
Try this:
FROM dbo.SYS_COMP_ADDRESS
CROSS JOIN dbo.SL_ACCOUNTS
INNER JOIN dbo.SL_TRANSACTIONS
INNER JOIN dbo.ORD_HEADER
ON dbo.SL_TRANSACTIONS.ST_ORDER_NUMBER =
dbo.ORD_HEADER.OH_ORDER_NUMBER
INNER JOIN dbo.SL_PL_NL_DETAIL
ON dbo.SL_TRANSACTIONS.ST_HEADER_KEY =
dbo.SL_PL_NL_DETAIL.DET_HEADER_KEY
LEFT OUTER JOIN dbo.STK_STOCK
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK.STKCODE
INNER JOIN dbo.ORD_DETAIL
ON dbo.SL_PL_NL_DETAIL.DET_ORDER_LINK =
dbo.ORD_DETAIL.OD_PRIMARY
LEFT OUTER JOIN dbo.STK_STOCK3
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK3.STKCODE3
LEFT OUTER JOIN dbo.STK_STOCK_2
ON dbo.SL_PL_NL_DETAIL.DET_STOCK_CODE =
dbo.STK_STOCK_2.STKCODE2
LEFT OUTER JOIN dbo.PRC_PRICE_RECS2
ON dbo.SL_PL_NL_DETAIL.DET_PRICE_CODE =
dbo.PRC_PRICE_RECS2.PRCODE2
ON dbo.SL_ACCOUNTS.CUCODE = dbo.SL_TRANSACTIONS.ST_COPYCUST
INNER JOIN dbo.SL_ACCOUNTS2
ON dbo.SL_ACCOUNTS.CUCODE = dbo.SL_ACCOUNTS2.CUCODE2
WHERE ( dbo.SL_TRANSACTIONS.ST_TRANTYPE IN ( 'INV', 'CRN' ) )
AND ( dbo.SL_TRANSACTIONS.ST_BATCH_FLAG <> 1 )
AND ( dbo.SL_ACCOUNTS.CUCODE <> '1023134' )
AND ( dbo.SL_ACCOUNTS.CUCODE <> '1023265' )
AND ( dbo.SL_ACCOUNTS.CUSORT NOT LIKE '%other income%' )
AND ( dbo.SL_ACCOUNTS.CUSORT NOT LIKE '%supplier%' )
Are you using the old style join syntax anywhere so search for *= and =* in your view definition
After applying join condition on two tables I want records which is maximum among records of left table
My query
SELECT a1.*,
t.*,
( a1.trnratefrom - t.trnratefrom )AS minrate,
( a1.trnrateto - t.trnrateto ) AS maxrate
FROM (SELECT a.srno,
trndate,
b.trnsrno,
Upper(Rtrim(Ltrim(b.trnstate))) AS trnstate,
Upper(Rtrim(Ltrim(b.trnarea))) AS trnarea,
Upper(Rtrim(Ltrim(b.trnquality))) AS trnquality,
Upper(Rtrim(Ltrim(b.trnlength))) AS trnlength,
Upper(Rtrim(Ltrim(b.trnunit))) AS trnunit,
b.trnratefrom,
b.trnrateto,
a.remark,
entdate
FROM mstprodrates a
INNER JOIN trnprodrates b
ON a.srno = b.srno)a1
INNER JOIN (SELECT c.srno,
trndate,
d.trnsrno,
Upper(Rtrim(Ltrim(d.trnstate))) AS trnstate,
Upper(Rtrim(Ltrim(d.trnarea))) AS trnarea,
Upper(Rtrim(Ltrim(d.trnquality))) AS trnquality,
Upper(Rtrim(Ltrim(d.trnlength))) AS trnlength,
Upper(Rtrim(Ltrim(d.trnunit))) AS trnunit,
d.trnratefrom,
d.trnrateto,
c.remark,
entdate
FROM mstprodrates c
INNER JOIN trnprodrates d
ON c.srno = d.srno) AS t
ON a1.trnstate = t.trnstate
AND a1.trnquality = t.trnquality
AND a1.trnunit = t.trnunit
AND a1.trnlength = t.trnlength
AND a1.trnarea = t.trnarea
AND a1.remark = t.remark
WHERE t.srno = (SELECT MAX(srno)
FROM a1
WHERE srno < a1.srno)
If you mean to say,
you want Records exist in Left table but not in right then use LEFT OUTER JOIN.