休眠查询性能

Question

我决定在现有应用程序上测试 nhibernate（将 linq 替换为 sql），到目前为止我看到的结果很糟糕。希望这里有人能指出我正确的方向。

我的配置：

<hibernate-configuration xmlns="urn:nhibernate-configuration-2.2">
    <session-factory>
        <property name="dialect">NHibernate.Dialect.MsSql2000Dialect</property>
        <property name="connection.provider">NHibernate.Connection.DriverConnectionProvider</property>
        <property name="connection.driver_class">NHibernate.Driver.SqlClientDriver</property>
        <property name="connection.connection_string">{my connection string here}</property>
        <property name="proxyfactory.factory_class">NHibernate.ByteCode.LinFu.ProxyFactoryFactory, NHibernate.ByteCode.LinFu</property>
        <property name="adonet.batch_size">10</property>
        <property name="show_sql">false</property>           
        <property name="use_outer_join">true</property>
        <property name="command_timeout">60</property>
        <property name="query.substitutions">false</property>            
    </session-factory>
</hibernate-configuration>

领域类：

public class Code
{
    public virtual Guid CodeId { get; set; }
    public virtual string CodeValue { get; set; }
    public virtual Guid EntryId { get; set; }
}

映射xml：

<?xml version="1.0" encoding="utf-8" ?>
<hibernate-mapping xmlns="urn:nhibernate-mapping-2.2" assembly="SweepstakesWeb" namespace="SweepstakesWeb.Domain">
    <class name="Code" table="Codes">
        <id name="CodeId">
            <generator class="guid.comb"/>
        </id>
        <property name="CodeValue" length="20" />
        <property name="EntryId" />
    </class>
</hibernate-mapping>

当我运行测试时，我看到了这个：

对于以下查询：

SELECT this_.CodeId as CodeId0_0_, this_.CodeValue as CodeValue0_0_, this_.EntryId as EntryId0_0_ FROM Codes this_ WHERE this_.CodeValue = N'3734872FVD'

我明白了：

StmtText
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  |--Nested Loops(Inner Join, OUTER REFERENCES:([this_].[CodeID]) OPTIMIZED)
       |--Index Scan(OBJECT:([Codes].[ix_codes_code] AS [this_]),  WHERE:(CONVERT_IMPLICIT(nvarchar(20),[Codes].[CodeValue] as [this_].[CodeValue],0)=N'3734872FVD'))
       |--Clustered Index Seek(OBJECT:([Codes].[PK__Codes__3493CFA7] AS [this_]), SEEK:([this_].[CodeID]=[Codes].[CodeID] as [this_].[CodeID]) LOOKUP ORDERED FORWARD)

如果我像这样手动将参数更改为 varchar（删除 N 前缀）：

SELECT this_.CodeId as CodeId0_0_, this_.CodeValue as CodeValue0_0_, this_.EntryId as EntryId0_0_ FROM Codes this_ WHERE this_.CodeValue = '3734872FVD'

我明白了：

StmtText
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  |--Nested Loops(Inner Join, OUTER REFERENCES:([this_].[CodeID]) OPTIMIZED)
       |--Index Seek(OBJECT:([Codes].[ix_codes_code] AS [this_]), SEEK:([this_].[CodeValue]='3734872FVD') ORDERED FORWARD)
       |--Clustered Index Seek(OBJECT:([Codes].[PK__Codes__3493CFA7] AS [this_]), SEEK:([this_].[CodeID]=[Codes].[CodeID] as [this_].[CodeID]) LOOKUP ORDERED FORWARD)

请注意，使用 varchar 参数值而不是 nvarchar，我会得到一个索引搜索。使用 nvarchar，我得到一个索引扫描。列类型是 varchar(20)。显然，正在进行的隐式转换导致完全扫描而不是搜索。

1. 为什么隐式转换需要扫描而不是查找？
2. 为什么 nhibernate 默认使用 nvarchar，如何告诉它使用 varchar？

Answer 1

该死。发帖几分钟后，我找到了答案：

http://community.devpinoy.org/blogs/joeycalisay/archive/2007/05/11/nhibernate-unicodes-and-index.aspx