Assistance with Query

Question

All...
I'm looking for assistance on a query that is running for a significant period of time - I have taken this query over from a colleague who is no longer available, I am relatively new with SQL  but a quick study - I have made the adjustment to the Stored
Procedure to Update the transactions in the attached SQL and the output works just fine...  the challenge is when I run this against large volumes of data it takes an unacceptable amount of time...  How can I streamline this code and achieve the
same outcome?
Stored Procedure extract -
declare @tb_NewKey table
 AccountNumber varchar(9) ,
 InstNumber int ,
 InstDueDate smalldatetime,
 ActionFlg varchar(20),
 OldHashKey varchar(50),
 NewHashKey varchar(50),
 DataSourcedate smalldatetime
insert into @tb_NewKey
 AccountNumber,
 InstNumber,
 InstDueDate,
 ActionFlg,
 OldHashKey,
 NewHashKey,
 DataSourcedate
select
 a.AccountNumber,
 a.InstNumber,
 a.InstDueDate,
 case when s.AccountNumber is null then 'Insert'
      when a.HashKey = s.HashKey then 'Skip'
      else 'Update' end ActionFlg,
 s.HashKey OldHashKey,
 a.HashKey NewHashKey,  
 convert(varchar(10),getdate(),126) DataSourceDate
from
 select
  a.AccountNumber,
  a.InstNumber,
  a.InstDueDate,
   upper(master.dbo.fn_varbintohexstr(hashbytes('MD5',
   CASE WHEN a.GroupNumber is null THEN '' ELSE a.GroupNumber END+
   CASE WHEN a.InstMailDate is null THEN ''ELSE convert(varchar(50),a.InstMailDate) END+
   CASE WHEN a.InstDueDate is null THEN ''ELSE convert(varchar(50),a.InstDueDate) END+
   STR(a.InstAmount)
   ))) HashKey   --
 from dbo.r_Billing_ScheduleInfo a
) a
left outer join dbo.stg_Billing_ScheduleInfo s
on (a.AccountNumber = s.AccountNumber and a.InstNumber = s.InstNumber)
With U as
 select tb.* from @tb_NewKey tb where tb.accountnumber in (select distinct accountnumber from @tb_NewKey where ActionFlg = 'Update')
 Update U
 Set ActionFlg = 'Update';
The SQL Stored procedure task is performing the following:
Declare a Temporary table
Insert transactions from a new input source table JOINED with data rows from the existing data table into the Temporary Table...  The transactions inserted into the Temporary Table can be brand new transactions or can be existing transactions from
the JOIN
- Example - If any of the input data source rows represents updated data for an existing row it is marked as an Update, then all Existing rows JOINED into the Temporary Table should be marked for Update - This step subsequently updates a real data
table
NEW DATA SOURCE -
200000705 1 2010-02-10  Update    
200000705 2 2010-03-10  Update
200000705 3 2010-04-10  Skip
200000705 4 2010-05-10  Skip
EXPECTED OUTCOME
200000705 1 2010-02-10  Update    
200000705 2 2010-03-10  Update
200000705 3 2010-04-10  Update
200000705 4 2010-05-10  Update

Answer 1

To be brutally honest, I don't have the time to figure my way through that web of.. stuff.
However, I don't want to leave you hanging totally, so here's a simplified version of the concept which might help but you down the right path:
declare @targetTable TABLE (id INT, date DATE)
INSERT INTO @targetTable (id, date)
VALUES
(1,GETDATE()-4),(2,GETDATE()-3),(3,GETDATE()-2),(4,GETDATE()-1)
DECLARE @sourceTable TABLE (id INT, date DATE)
INSERT INTO @sourceTable (id, date)
VALUES
(1,GETDATE()),(2,GETDATE()),(2,GETDATE()-3),(3,GETDATE()-2),(5,GETDATE()),(6,GETDATE()+1)
SELECT *,
CASE
WHEN tt.id IS NOT NULL AND tt.date <> st.date THEN 'UPDATE'
WHEN tt.id IS NOT NULL AND tt.date = st.date THEN 'SKIP'
WHEN tt.id IS NULL THEN 'INSERT'
END AS actionFlag
FROM @sourceTable st
LEFT OUTER JOIN @targetTable tt
ON st.id = tt.id