"ADO. net-Intermediate "Two ways to test millions data in bulk INSERT

Source: Internet
Author: User
Tags bulk insert connectionstrings

Inserting a single piece of data into SQL Server uses the INSERT statement, but if you want to bulk insert a bunch of data, looping through insert is inefficient and results in a system performance problem with SQL. The following are the two bulk data insertion methods supported by SQL Server: Bulk and table-valued parameters (table-valued Parameters).

Run the following script to establish the test database and table-valued parameters.

-CREATE DATABASE  CREATE database bulktestdb;  Go use  bulktestdb;  Go  --CREATE TABLE  CREATE TABLE bulktesttable (  int  primary key,  UserName nvarchar (+),  Pwd varchar (+))  go  -- Create Table Valued  CREATE TYPE Bulkudt as TABLE    int,     UserName nvarchar (+),     Pwd varchar (+))  

Let's insert 1 million data using the simplest INSERT statement, with the following code:

Stopwatch SW =NewStopwatch (); SqlConnection sqlconn=NewSqlConnection (configurationmanager.connectionstrings["ConnStr"]. ConnectionString);//connecting to a databaseSqlCommand Sqlcomm=NewSqlCommand (); Sqlcomm.commandtext=string. Format ("INSERT INTO bulktesttable (ID,USERNAME,PWD) VALUES (@p0, @p1, @p2)");//Parameterized SQLSQLCOMM.PARAMETERS.ADD ("@p0", SqlDbType.Int); SQLCOMM.PARAMETERS.ADD ("@p1", SqlDbType.NVarChar); SQLCOMM.PARAMETERS.ADD ("@p2", SqlDbType.VarChar); Sqlcomm.commandtype=CommandType.Text; Sqlcomm.connection=sqlconn;  Sqlconn.open (); Try  {      //the loop inserts 1 million data, inserts 100,000, inserts 10 times each time.      for(intMultiply =0; Multiply <Ten; multiply++)      {           for(intCount = Multiply *100000; Count < (multiply +1) *100000; count++) {sqlcomm.parameters["@p0"]. Value =count; sqlcomm.parameters["@p1"]. Value =string. Format ("user-{0}", Count *multiply); sqlcomm.parameters["@p2"]. Value =string. Format ("pwd-{0}", Count *multiply); Sw.              Start ();              Sqlcomm.executenonquery (); Sw.          Stop (); }          //when 100,000 data is inserted, the time taken for this insertion is displayedConsole.WriteLine (string. Format ("Elapsed time is {0} Milliseconds", SW.      Elapsedmilliseconds)); }  }  Catch(Exception ex) {Throwex; }  finally{sqlconn.close ();  } console.readline (); 

The time-consuming diagram is as follows:

Since the operation is too slow to insert 100,000 is time-consuming 72390 milliseconds, so I manually forced to stop.

Here's a look at the use of bulk inserts:

The main idea of the bulk method is to insert the data in the table into the database by sqlbulkcopy the data in the table at the client, and then using the

The code is as follows:

 Public Static voidBulktodb (DataTable dt) {SqlConnection sqlconn=NewSqlConnection (configurationmanager.connectionstrings["ConnStr"].      ConnectionString); SqlBulkCopy bulkcopy=NewSqlBulkCopy (sqlconn); Bulkcopy.destinationtablename="bulktesttable"; Bulkcopy.batchsize=dt.        Rows.Count; Try{sqlconn.open (); if(dt! =NULL&& dt. Rows.Count! =0) bulkcopy.writetoserver (DT); }      Catch(Exception ex) {Throwex; }      finally{sqlconn.close (); if(BulkCopy! =NULL) Bulkcopy.close (); }  }     Public StaticDataTable GetTableSchema () {DataTable dt=NewDataTable (); Dt. Columns.addrange (Newdatacolumn[]{NewDataColumn ("Id",typeof(int)),          NewDataColumn ("UserName",typeof(string)),      NewDataColumn ("PWD",typeof(string))}); returnDT; }    Static voidMain (string[] args) {Stopwatch SW=NewStopwatch ();  for(intMultiply =0; Multiply <Ten; multiply++) {DataTable dt=Bulk.gettableschema ();  for(intCount = Multiply *100000; Count < (multiply +1) *100000; count++) {DataRow R=dt.              NewRow (); r[0] =count; r[1] =string. Format ("user-{0}", Count *multiply); r[2] =string. Format ("pwd-{0}", Count *multiply); Dt.          Rows.Add (R); } SW.          Start ();          BULK.BULKTODB (DT); Sw.          Stop (); Console.WriteLine (string. Format ("Elapsed time is {0} Milliseconds", SW.      Elapsedmilliseconds));  } console.readline (); }  

The time-consuming diagram is as follows:

It can be seen that efficiency and performance increase significantly after using bulk. Inserting 100,000 data with insert takes 72390, and now it takes 17583 to insert 1 million data using bulk.

Finally, let's look at the efficiency of using table-valued parameters and you'll be surprised.

The table-valued parameter is a new SQL Server 2008 feature, referred to as TVPs. For friends who are not familiar with table-valued parameters, you can refer to the latest book online, I will also write a blog about table-valued parameters, but this time does not introduce the concept of table-valued parameters too much. Anyway, look at the code:

[C-Sharp ] View plain copy Public Static voidTablevaluedtodb (DataTable dt) {SqlConnection sqlconn=NewSqlConnection (configurationmanager.connectionstrings["ConnStr"].      ConnectionString); Const stringTsqlstatement ="INSERT INTO bulktesttable (ID,USERNAME,PWD)"+"SELECT NC. Id, NC. Username,nc. PWD"+"From @NewBulkTestTvp as NC"; SqlCommand cmd=NewSqlCommand (tsqlstatement, sqlconn); SqlParameter Catparam= cmd. Parameters.addwithvalue ("@NewBulkTestTvp", DT); Catparam.sqldbtype=sqldbtype.structured; //The name of the table-valued parameter is Bulkudt, which is in the SQL that created the test environment above. Catparam.typename ="dbo. Bulkudt"; Try{sqlconn.open (); if(dt! =NULL&& dt. Rows.Count! =0) {cmd.        ExecuteNonQuery (); }      }      Catch(Exception ex) {Throwex; }      finally{sqlconn.close (); }  }     Public StaticDataTable GetTableSchema () {DataTable dt=NewDataTable (); Dt. Columns.addrange (Newdatacolumn[]{NewDataColumn ("Id",typeof(int)),        NewDataColumn ("UserName",typeof(string)),        NewDataColumn ("PWD",typeof(string))}); returnDT; }    Static voidMain (string[] args) {Stopwatch SW=NewStopwatch ();  for(intMultiply =0; Multiply <Ten; multiply++) {DataTable dt=Tablevalued.gettableschema ();  for(intCount = Multiply *100000; Count < (multiply +1) *100000; count++) {DataRow R=dt.              NewRow (); r[0] =count; r[1] =string. Format ("user-{0}", Count *multiply); r[2] =string. Format ("pwd-{0}", Count *multiply); Dt.          Rows.Add (R); } SW.          Start ();          TABLEVALUED.TABLEVALUEDTODB (DT); Sw.          Stop (); Console.WriteLine (string. Format ("Elapsed time is {0} Milliseconds", SW.      Elapsedmilliseconds));  } console.readline (); }  

The time-consuming diagram is as follows:

It's 5 seconds faster than bulk.

"ADO. net-Intermediate "Two ways to test millions data in bulk INSERT

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.