测试环境：

   springframework 1.0

       数据库：Informix 9.21.FC1

       表结构：

       Column name          Type                                    Nulls

f1                   varchar(10,0)                           yes

f2                   varchar(10,0)                           yes

f3                   varchar(10,0)                           yes

f4                   varchar(10,0)                           yes

f5                   varchar(10,0)                           yes

f6                   varchar(10,0)                           yes

f7                   varchar(10,0)                           yes

f8                   varchar(10,0)                           yes

f9                   varchar(10,0)                           yes

f10                  varchar(10,0)                           yes

在f1上建立索引

测试情况：

case 1：SQL单个提交，不prepare，使用BaseDAOJdbc的executeSql方法

case 2：SQL单个提交，prepare，方法如下：

public void saveDataSinglePrepare(final List data) {

        getJdbcTemplate().execute(insertData,

            new PreparedStatementCallback() {

                          public Object doInPreparedStatement(PreparedStatement ps) throws SQLException, DataAccessException {

                              Iterator iter = data.iterator();

                              while (iter.hasNext()){

                                  Object[] objs = (Object[]) iter.next();

                                  for(int i=0; i

                                            ps.setObject(i+1, objs[i]);

                                           }

                                           ps.execute();

                              }

                              return null;

                          }

                }

    );

    }

case 3：SQL批量提交，使用BaseDAOJdbc的executeBatch方法

/**

 * Jdbc sql的批量操作

 * @param sql       sql语句

 * @param values 包含参数的List，List的每个元素是一个Object[]，Object[]是一个单独的sql操作所需的参数

 */

public void executeBatch(final String sql, final List values){

    if (values.size() == 0) return ;

    getJdbcTemplate().batchUpdate(sql,

        new BatchPreparedStatementSetter() {

               public void setValues(PreparedStatement ps, int i) throws SQLException {

                   Object[] args = (Object[]) values.get(i);

                   for(int j=0; j

                       ///log.debug(String.valueOf(j+1)+":"+args[j].toString());

                              ps.setObject(j+1, args[j]);

                             }

            }

            public int getBatchSize() {

                return values.size();

            }

               }

    );  

}

case 4：SQL批量提交，使用BaseDAOJdbc的executeBatch2方法

private static final int batch_size = 100;

public void executeBatch2(final String sql, final List values){

    getJdbcTemplate().execute(sql,

        new PreparedStatementCallback() {

                      public Object doInPreparedStatement(PreparedStatement ps) throws SQLException, DataAccessException {

                          Iterator iter = values.iterator();

                          int sqlCount = 0;

                          while (iter.hasNext()){

                              if (sqlCount >= batch_size){

                                  ps.executeBatch();

                                  sqlCount = 0;

                                  //log.info("executeBatch:" + sql);

                              }

                              Object[] args = (Object[]) iter.next();

                              for(int i=0; i

                                     ps.setObject(i+1, args[i]);

                                    }

                              ps.addBatch();

                           sqlCount ++;

                           log.debug("sqlCount:" + String.valueOf(sqlCount));

                          }

                          if (sqlCount > 0){

                           ps.executeBatch();

                           sqlCount = 0;

                           //log.info("executeBatch:" + sql);

                       }

                             return null;

                      }

        }

    );  

}