In This Topic
Array binding is a powerful feature that significantly improves the performance by executing similar SQL statements in a single batch, rather than sending them one by one to the database.
The feature is especially useful for performing batch operations of the same type, such as INSERT, UPDATE, or DELETE.
The article demonstrates batch data insertion and update using DB2Command with a practical example. To update your database with large volumes of data, follow the steps below:
1. Create a database and its table.
With any SQL tool, you can create a database and its table by using the following DDL statement:
CREATE TABLE batch_test(" +
"ID INTEGER NOT NULL, " +
"F_INTEGER INTEGER, " +
"F_VARCHAR VARCHAR(100 OCTETS), " +
"PRIMARY KEY(ID)
)";
Alternatively, you can achieve the same result programmatically with the following C# code:
using(DB2Connection connection = new
DB2Connection("Server=127.0.0.1:50000;Database=SAMPLE1;uid=test;pwd=test;currentSchema=test;"))
{
connection.Open();
// create the test table
DB2Command createCommand = new DB2Command("CREATE TABLE batch_test(" +
"ID INTEGER NOT NULL, " +
"F_INTEGER INTEGER, " +
"F_VARCHAR VARCHAR(100 OCTETS), " +
"PRIMARY KEY(ID))",
connection);
createCommand.ExecuteNonQuery();
}
2. Create DB2Command and set a DML statement.
Define the desired DML statement (e.g. INSERT, UPDATE, or DELETE) by setting the CommandText property of the DB2Command object.
DB2Command command = new DB2Command("INSERT INTO batch_test" +
"(ID, F_INTENGER, F_VARCHAR) " +
"VALUES" +
"(:ID, :F_INTENGER, :F_VARCHAR)",
connection);
3. Add parameters.
Specify the parameter names and their types.
command.Parameters.Add("id", DB2Type.Integer, 4, "id");
command.Parameters.Add("f_integer", DB2Type.Integer, 4, "f_integer");
command.Parameters.Add("f_varchar", DB2Type.VarChar, 100, "f_varchar");
Array binding with DB2Command doesn't support explicitly setting the batch size. The optimal size is calculated automatically.
4. Fill parameter values.
Define an array of values in each parameter of the DB2Command. Each array element corresponds to a different execution of the SQL statement within the same batch.
command.Parameters["id"].Value = new int[5] { 1, 2, 3, 4, 5};
command.Parameters["f_integer"].Value = new int[5] { 1, 2, 3, 4, 5 };
command.Parameters["f_varchar"].Value = new string[5] { "string 1", "string 2", "string 3", "string 4", "string 5" };
5. Execute the batch insert.
Call the ExecuteArray() method to execute the command for all sets of parameter values in a single batch.
Below is a sample code that executes several INSERT operations using array binding.
CREATE TABLE batch_test(" +
"ID INTEGER NOT NULL, " +
"F_INTEGER INTEGER, " +
"F_VARCHAR VARCHAR(100 OCTETS), " +
"PRIMARY KEY(ID)
)";
using (DB2Connection connection = new
DB2Connection("Server=127.0.0.1:50000;Database=SAMPLE1;uid=test;pwd=test;currentSchema=test;"))
{
connection.Open();
// create DB2Command
DB2Command command = new DB2Command("INSERT INTO batch_test" +
"(id, f_integer, f_varchar) " +
"VALUES" +
"(:id, :f_integer, :f_varchar)",
connection);
command.Parameters.Add("id", DB2Type.Integer, 4, "id");
command.Parameters.Add("f_integer", DB2Type.Integer, 4, "f_integer");
command.Parameters.Add("f_varchar", DB2Type.VarChar, 100, "f_varchar");
// fill DB2Command parameter values
command.Parameters["id"].Value = new int[5] { 1, 2, 3, 4, 5};
command.Parameters["f_integer"].Value = new int[5] { 1, 2, 3, 4, 5 };
command.Parameters["f_varchar"].Value = new string[5] { "string 1", "string 2", "string 3", "string 4", "string 5" };
// execute the update
command.ExecuteArray();
}
This example demonstrates how to insert multiple records with a single execution operation using array binding in dotConnect for DB2.
To use the array binding feature, you must assign arrays of values to the parameters of a DB2Command object and call the ExecuteArray() method. Each array element corresponds to a different execution of the SQL statement within the same batch.
Array binding with DB2Command doesn't support explicitly setting the batch size. The optimal size is calculated automatically.
This approach is especially beneficial when working with large datasets, as it reduces the number of individual communications between the application and the database, significantly improving performance.