//方法一：用DataAdapter方法，优点：不用写insert语句且不用关心values后边的各字段的形式，如要不要加单引号之类的；缺点:插入一条记录的效率不及写单纯insert语句SqlConnection conn = ConnectDB(server, db);string sql = "select * from " + tableName;SqlDataAdapter da = new SqlDataAdapter(sql, conn);SqlCommandBuilder scb = new SqlCommandBuilder(da);//SqlCommandBuilder根据insertCommand构造    updatecommand和deletecommandDataTable dt = new DataTable();exeTable(conn, dt, sql);DataRow dr = dt.NewRow(); dr.ItemArray = columnValues; dt.Rows.Add(dr); da.Update(dt); conn.Close(); // 方法二：单纯写insert语句插入一条记录，优点:单条数据插入效率高 缺点：需要注意sql语句的构造和sql语句中参数的类型 string sql = "insert into 表名 values(" + 值1+ "," + 值2+ ",'" + 值3+ "',"+ 值4+ "," + 值5+ "," + 值6+",@参数1,'" + 值7+ "','"+ 值8+ "')"; SqlParameter param = new System.Data.SqlClient.SqlParameter("@参数1", SqlDbType.Image); param.Value = 值9; SqlConnection conn = sqlHelper.ConnectDB(); SqlCommand cmd = new System.Data.SqlClient.SqlCommand(sql, conn); if (param != null) cmd.Parameters.Add(param); cmd.ExecuteNonQuery(); conn.Close(); //方法三:插入大批量数据时用SqlBulkCopy，单条记录插入效率较慢  DataTable dt = new DataTable(); dt.Columns.Add("列名1", typeof(string)); dt.Columns.Add("列名2", typeof(byte[])); DataRow dr = dt.NewRow(); dr["列名1"] = 值1; dr["列名2"] = 值2; dt.Rows.Add(dr); dr = dt.NewRow(); dr["列名1"] = 值3; dr["列名2"] = 值4; dt.Rows.Add(dr); SqlBulkCopy bulk = new System.Data.SqlClient.SqlBulkCopy(conn); bulk.DestinationTableName=目标表名; bulk.BatchSize=dt.Rows.Count; bulk.ColumnMappings.Add("列名1"，目标列名1); bulk.ColumnMappings.Add("列名2"，目标列名2); bulk.WriteToServer(dt); bulk.Close(); conn.Close();