Linq to sql 简单性能差异指引

测试环境：

cpu: tk55

ram: 2g

hd:120g

os:vista+sp1

sql:sql2005开发版

以下测试只跑一两次，取最后一次（一般第一次会很慢）。

1.禁用ObjectTrackingEnabled

当只是查询数据而不跟踪对象状态时禁用，将极大的提供查询性能。性能差异在36倍左右。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

adventureWorks1.ObjectTrackingEnabled = true;

var q1 = (from p in adventureWorks1.Products select p).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

adventureWorks2.ObjectTrackingEnabled = false;

var q2 = (from p in adventureWorks2.Products select p).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
输出：

11232000

312000

2.不记录输出

输入记录一般只用在开发阶段，在部署环境中一般不使用。性能差异在17-18倍左右。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

adventureWorks1.Log = Console.Out;

var q1 = (from p in adventureWorks1.Products select p).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

var q2 = (from p in adventureWorks2.Products select p).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);

11076000

624000

3.编译lambda表达式

性能差异14-15倍。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

var q1 = (from p in adventureWorks1.Products select p).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

var compiledQuery = CompiledQuery.Compile((AdventureWorksDataContext ctx) => from p in ctx.Products select p);

var q2 = compiledQuery(adventureWorks2).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
输出：

11076000

780000

4.使用DataLoadOptions.LoadWith

性能差异5-6倍左右。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

var q1 = (from o in adventureWorks1.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);
AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

DataLoadOptions options = new DataLoadOptions();

options.LoadWith<WorkOrder>(o => o.Product);

adventureWorks2.LoadOptions = options;

var q2 = (from o in adventureWorks2.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);

输出：

11700000

2184000

5.使用延迟加载(系统默认)

性能差异42-43倍左右。测试代码
AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

adventureWorks1.DeferredLoadingEnabled = true;

var q1 = (from o in adventureWorks1.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

adventureWorks2.DeferredLoadingEnabled = false;

var q2 = (from o in adventureWorks1.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
输出：

26676000

624000

6.使用sql语句

性能差异 7倍左右
AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

var q1 = (from o in adventureWorks1.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

var q2 = adventureWorks2.ExecuteQuery<Product>(@"SELECT [t0].[WorkOrderID], [t0].[ProductID], [t0].[OrderQty], [t0].[StockedQty], [t0].[ScrappedQty], [t0].[StartDate], [t0].[EndDate], [t0].[DueDate], [t0].[ScrapReasonID], [t0].[ModifiedDate]

FROM [Production].[WorkOrder] AS [t0]

WHERE [t0].[ProductID] = 3").ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
输出：

11856000

1716000

7.使用储存过程

性能差异20倍左右。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

var q1 = (from o in adventureWorks1.WorkOrders where o.ProductID == 3 select o).ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

var q2 = adventureWorks2.GetWorkOrderByProductID(3).ToList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
-- ================================================

-- Template generated from Template Explorer using:

-- Create Procedure (New Menu).SQL

--

-- Use the Specify Values for Template Parameters

-- command (Ctrl-Shift-M) to fill in the parameter

-- values below.

--

-- This block of comments will not be included in

-- the definition of the procedure.

-- ================================================

SET ANSI_NULLS ON

GO

SET QUOTED_IDENTIFIER ON

GO

-- =============================================

-- Author: <Author,,Name>

-- Create date: <Create Date,,>

-- Description: <Description,,>

-- =============================================

CREATE PROCEDURE Production.GetWorkOrderByProductID

-- Add the parameters for the stored procedure here

@ProductID int

AS

BEGIN

-- SET NOCOUNT ON added to prevent extra result sets from

-- interfering with SELECT statements.

SET NOCOUNT ON;

SELECT [t0].[WorkOrderID], [t0].[ProductID], [t0].[OrderQty], [t0].[StockedQty], [t0].[ScrappedQty], [t0].[StartDate], [t0].[EndDate], [t0].[DueDate], [t0].[ScrapReasonID], [t0].[ModifiedDate]

FROM [Production].[WorkOrder] AS [t0]

WHERE [t0].[ProductID] = @ProductID

END

GO

输出：

12168000

624000

8、获取数据以用于数据绑定

性能差异1.5倍左右。测试代码：

AdventureWorksDataContext adventureWorks1 = new AdventureWorksDataContext();

DateTime d1 = DateTime.Now;

var q1 = adventureWorks1.Products.ToList();

DateTime d2 = DateTime.Now;

Console.WriteLine(d2.Ticks - d1.Ticks);

AdventureWorksDataContext adventureWorks2 = new AdventureWorksDataContext();

DateTime d3 = DateTime.Now;

var q2 = adventureWorks2.WorkOrders.GetNewBindingList();

DateTime d4 = DateTime.Now;

Console.WriteLine(d4.Ticks - d3.Ticks);
输出：

10140000

6708000