在为应用程序调用表单 ASP .NET Windows 时，如何优化我的 Sql 查询？

Question

There are many sub query's in my query. When I execute the query in Sql Server Management studio it doesn't take much time. But when I call this query in my ASP .NET Windows for Application then I get the following error.

System.Data.SqlClient.SqlException: 'Execution Timeout Expired. The timeout period elapsed prior to completion of the operation or the server is not responding.'

My query is:

SELECT
  id.Id,
  id.CardNo,
  (CASE
    WHEN id.NameTitle = 0 THEN ''
    ELSE CASE
        WHEN id.NameTitle = 1 THEN 'Mr. '
        ELSE CASE
            WHEN id.NameTitle = 2 THEN 'Mrs. '
            ELSE CASE
                WHEN id.NameTitle
                  = 3 THEN 'Md. '
                ELSE CASE
                    WHEN id.NameTitle = 4 THEN 'Engr. '
                    ELSE CASE
                        WHEN id.NameTitle = 5 THEN 'Dr. '
                        ELSE CASE
                            WHEN id.NameTitle = 6 THEN 'Mosa. '
                            ELSE CASE
                                WHEN id.NameTitle = 7 THEN 'Shree. '
                                ELSE CASE
                                    WHEN id.NameTitle = 8 THEN 'Mst. '
                                    ELSE CASE
                                        WHEN id.NameTitle = 9 THEN 'Miss. '
                                        ELSE ''
                                      END
                                  END
                              END
                          END
                      END
                  END
              END
          END
      END
  END) + (firstName.[Name] + ' ' + secondName.Name + ' ' + thirdName.Name + ' ' + fourthame.Name) AS Name
FROM dbo.EmployeeDatas id
JOIN dbo.EmployeeNames firstName
  ON id.Id = firstName.EmployeeDataId
JOIN dbo.EmployeeNames secondName
  ON id.Id = secondName.EmployeeDataId
JOIN dbo.EmployeeNames thirdName
  ON id.Id = thirdName.EmployeeDataId
JOIN dbo.EmployeeNames fourthame
  ON id.Id = fourthame.EmployeeDataId
WHERE firstName.DataTypeId = 1
AND secondName.DataTypeId = 2
AND thirdName.DataTypeId = 3
AND fourthame.DataTypeId = 4

Sql Server Execution

My Cs file in Windows Application is:

namespace SnowtexBiometricExe.Forms
{
    public partial class frmRegistration : Form
    {
        public frmRegistration()
        {
            InitializeComponent();
        }
        private void FrmRegistration_Load(object sender, EventArgs e)
        {
            string mainConn = ConfigurationManager.ConnectionStrings["SBR"].ConnectionString;
            SqlConnection connection = new SqlConnection(mainConn);
            string query = "select id.Id, id.CardNo, (CASE WHEN id.NameTitle = 0 THEN '' ELSE CASE WHEN id.NameTitle = 1 THEN 'Mr. ' ELSE CASE WHEN id.NameTitle = 2 THEN 'Mrs. ' ELSE CASE WHEN id.NameTitle = 3 THEN 'Md. ' ELSE CASE WHEN id.NameTitle = 4 THEN 'Engr. ' ELSE CASE WHEN id.NameTitle = 5 THEN 'Dr. ' ELSE CASE WHEN id.NameTitle = 6 THEN 'Mosa. ' ELSE CASE WHEN id.NameTitle = 7 THEN 'Shree. ' ELSE CASE WHEN id.NameTitle = 8 THEN 'Mst. ' ELSE CASE WHEN id.NameTitle = 9 THEN 'Miss. ' ELSE '' END END END END END END END END END END) +(firstName.[Name] + ' ' + secondName.Name + ' ' + thirdName.Name + ' ' + fourthame.Name) AS Name  from dbo.EmployeeDatas id join dbo.EmployeeNames firstName on id.Id = firstName.EmployeeDataId join dbo.EmployeeNames secondName on id.Id = secondName.EmployeeDataId join dbo.EmployeeNames thirdName on id.Id = thirdName.EmployeeDataId join dbo.EmployeeNames fourthame on id.Id = fourthame.EmployeeDataId where firstName.DataTypeId = 1 and secondName.DataTypeId = 2 and thirdName.DataTypeId = 3 and fourthame.DataTypeId = 4";
            SqlCommand cmd = new SqlCommand(query, connection);
            connection.Open();
            SqlDataAdapter sda = new SqlDataAdapter(cmd);
            DataTable dataTable = new DataTable();
            sda.Fill(dataTable);
            lstEmployee.DisplayMember = "Name";
            lstEmployee.DataSource = dataTable;
            connection.Close();
        }        
    }
}

Error in windows application

Answer 1

you can set a timeout in your connection but i think its in your join that causing much delay.. can you provide the table structure of those joins?

and create a function instead of case

CREATE FUNCTION setTitle
(
@title int,
@firstname varchar(20),
@secondname varchar(20),
@thirdname varchar(20),
@fourthname varchar(20)
)
returns varchar(50)
as
begin
declare @myTitle as varchar(10);
declare @Name as varchar(50)
if(@title = 0)
set @myTitle = '';

else if(@title = 1)
set @myTitle = 'Mr.';

else if(@title = 2)
set @myTitle = 'Mrs.';

else if(@title = 3)
set @myTitle = 'Md.';

else if(@title = 4)
set @myTitle = 'Engr.';

else if(@title = 5)
set @myTitle = 'Dr.';

else if(@title = 6)
set @myTitle = 'Mosa.';

else if(@title = 7)
set @myTitle = 'Shree.';

else if(@title = 8)
set @myTitle = 'Mst.';

else if(@title = 9)
set @myTitle = 'Miss.';

select @Name = @myTitle + ' ' + @firstname + ' ' + @secondname  + ' ' + @thirdname + ' 
' + @fourthname 

return @Name
end

and in your query

SELECT
id.Id,
id.CardNo,
dbo.setTitle(id.NameTitle,firstName. 
[Name],secondName.Name,thirdName.Name,fourthame.Name) AS Name
FROM dbo.EmployeeDatas id
JOIN dbo.EmployeeNames firstName
ON id.Id = firstName.EmployeeDataId
JOIN dbo.EmployeeNames secondName
ON id.Id = secondName.EmployeeDataId
JOIN dbo.EmployeeNames thirdName
ON id.Id = thirdName.EmployeeDataId
JOIN dbo.EmployeeNames fourthame
ON id.Id = fourthame.EmployeeDataId
WHERE firstName.DataTypeId = 1
AND secondName.DataTypeId = 2
AND thirdName.DataTypeId = 3
AND fourthame.DataTypeId = 4

Answer 2

Note that I think you can simplify the CASE expressions greatly by making them into one CASE expression only eg,

SELECT
  id.Id,
  id.CardNo,
  (CASE id.NameTitle
      WHEN 0 THEN ''
      WHEN 1 THEN 'Mr. '
      WHEN 2 THEN 'Mrs. '
      WHEN 3 THEN 'Md. '
      WHEN 4 THEN 'Engr. '
      WHEN 5 THEN 'Dr. '
      WHEN 6 THEN 'Mosa. '
      WHEN 7 THEN 'Shree. '
      WHEN 8 THEN 'Mst. '
      WHEN 9 THEN 'Miss. '
      ELSE '' END)
    + (firstName.[Name] + ' ' + secondName.[Name] + ' ' + thirdName.[Name] + ' ' + fourthName.[Name]) AS [Name]
FROM dbo.EmployeeDatas id
    JOIN dbo.EmployeeNames firstName ON id.Id = firstName.EmployeeDataId
    JOIN dbo.EmployeeNames secondName ON id.Id = secondName.EmployeeDataId
    JOIN dbo.EmployeeNames thirdName ON id.Id = thirdName.EmployeeDataId
    JOIN dbo.EmployeeNames fourthName ON id.Id = fourthName.EmployeeDataId
WHERE firstName.DataTypeId = 1
    AND secondName.DataTypeId = 2
    AND thirdName.DataTypeId = 3
    AND fourthName.DataTypeId = 4;

You could also put it into a new table eg, Nametitle, with NametitleID and the actual NameTitle_short_text (for example) - and then using it in a LEFT JOIN or similar.

However, the above may not speed up your processing. It gives SQL Server a better shot at guessing what will happen, but it will all still be in-memory processing anyway.

---

For performance, however, I do suggest, a non-clustered index on dbo.EmployeeNames eg,

CREATE NONCLUSTERED INDEX IX_EmployeeNames_DataType ON dbo.EmployeeNames
    (DataTypeID, EmployeeDataId, Name);

This index will help with the 4 lookups you need to do - instead of reading the whole names table, it can seek to relevant spots.

Note that for small reads (eg, one or a few IDs) this will perform similarly or perhaps slightly worse than an index with the key columns the other way around eg, EmployeeDataId, DataTypeID, Name . However, I believe for larger result sets, this will give a better results.

It may be worth testing in your specific circumstances (note though, I'm guessing the Primary Key/clustered index for EmployeeNames is the second option above; if so then adding the same as a non-clustered index won't help).

---

Finally - note that your application/query has no WHERE clause - so you are getting all the data from these tables.

Is getting all the rows what you want? Or do you only want one (or a handful) of rows? If you can reduce the rows required, it will typically speed up processing by a large factor.