今天测试了一下 sqlalchemy 性能

self.db.query(Users).filter(Users.Id==1).first() < self.db.execute('SELECT *  FROM `users` WHERE Id=%s' % user_id).first() < self.db.queryRow('SELECT * FROM `users` WHERE `Id`=%s' % user_id)

sqlalchemy的echo=False的

ab测试如下：
ab -n 2000 -c 50 http://127.0.0.1:8888/

self.db.query(Users).filter(Users.Id==1).first()  的

Requests per second:    460.00左右 [#/sec] (mean)

self.db.execute('SELECT *  FROM `users` WHERE Id=%s' % user_id).first()

Requests per second:    860.00左右 [#/sec] (mean)

下面这个为 MySQLdb

self.db.queryRow('SELECT * FROM `users` WHERE `Id`=%s' % user_id)

Requests per second:    1110.00左右 [#/sec] (mean)

在想是否是用MySQLdb好呢还是用sqlalchemy好。

这MySQLdb我也改了下代码的：基本的CRUD如下：

    db = db(host='localhost', user='root', passwd='password', port=3306, db='test', charset='utf8');
    #row = db.queryRow('SELECT @@version')
    row = db.queryRow('SELECT   `id`,  `name` FROM `test` WHERE id=2')
    if(row):
        print row['name']
    else:
        print ''

    data = {}
    data['UserName'] = "mytest??"
    data['Email'] = "my2@test.com"
    data["Age"] = 26
    data["money"] = 6254.258
    #data['bc']="def"
    #data['flt']=38.5
    #添加数据
    #print db.execute_insert('mytb',data)
    #修改数据
    print db.execute_update('mytb', data, {'id':1, "Age":24})
    row = db.queryRow('select * from mytb where id = %s', (2))
    print row['UserName']

    for item in db.queryAll('select * from mytb'):
        print item['UserName']

据说 amysql  的性能比MySQLDb 快2.5倍。