Splitting a Large DataTable into Smaller Batches

One of the requirements that came in was to split a large datatable into smaller ones defined by a batch size (say 2000). The below mentioned sample code helps to achieve this easily. 

 

The given below method is used to create a large amount of sample data:

 

static DataTable GetTonsOfData()

{

   DataTable dt = new DataTable();

   dt.Columns.Add(new DataColumn("ID", typeof(Int32)));

   dt.Columns.Add(new DataColumn("Name", typeof(string)));

   for (int i = 1; i <= 20000; i++)

   {

        DataRow Row = dt.NewRow();

        Row["ID"] = i;

        Row["Name"] = "Name " + i.ToString();

        dt.Rows.Add(Row);

}

return dt;

}

 

The given below SplitTable method actually splits the original datatable into smaller datatables and insert them into a list of datatable.

 

private static List<DataTable> SplitTable(DataTable originalTable, int batchSize)

{

     List<DataTable> tables = new List<DataTable>();

     int i = 0;

     int j = 1;

    DataTable newDt = originalTable.Clone();

   newDt.TableName = "Table_" + j;

   newDt.Clear();

    foreach (DataRow row in originalTable.Rows)

    {

         DataRow newRow = newDt.NewRow();

         newRow.ItemArray = row.ItemArray;

         newDt.Rows.Add(newRow);

         i++;

         if (i == batchSize)

        {

           tables.Add(newDt);

           j++;

          newDt = originalTable.Clone();

          newDt.TableName = "Table_" + j;

          newDt.Clear();

          i = 0;

      }

  }

   return tables;

}

 

Now let us see the real working of the above methods.

 

DataTable dtFull = GetTonsOfData();

List<DataTable> splitdt = SplitTable(dtFull, 2000);

 

The code can be optimized if LINQ is implemented.