asp.net performance optimization.

Asp.net| Performance | optimization

(i). Select a session-state storage method
In webconfig file configuration:
<sessionstate mode= "???" Stateconnectionstring= "tcpip=127.0.0.1:42424"
sqlconnectionstring= "Data source=127.0.0.1; Trusted_connection=yes "
Cookieless= "false" timeout= "/>"

ASP.net has three ways of storing session-state information:
1. Stored in Process: Property mode = InProc
Features: Best performance, fastest speed, but not shared across multiple servers.

2. stored in State server: Property mode = "StateServer"
Features: Use this method when you need to maintain user session information across servers.
But the information is stored on the state server, and once the state server fails, the information is lost

3. stored in SQL Server: Properties mode= "SQL Server"
Feature: The workload will become larger, but the information will not be lost.

Add one point:
I. Because some pages do not require session state, you can disable session state:
The code is as follows: <%@ Page enablesessionstate= "false"%>
Ii. If the page requires access to session variables but does not allow them to be modified, you can set the page session state to read-only:
The code is as follows: <%@ Page enablesessionstate= "false"%>

When used, you can choose a certain way depending on the circumstances.
(ii). Use of Page.IsPostBack
Page.IsPostBack Indicates whether it was returned from the client. When it is first run, it is not returned from the client, its value
False, when the event on the page is triggered or when the page is refreshed, the value changes to true page.ispostback because it is postback;

Generally in: Page_Load method uses:
private void Page_Load (Object Sender,eventargs E)
{
if (! Page.IsPostBack)
{
....; Initializes the code for the page. The code executes the first time the page is initialized, and when the second postback,
Will not be executed again. Improve efficiency.
}
}

There are often times when you have to use IsPostBack, because some controls are initialized to maintain their state.
For example: DropDownList, if you initialize each time, the user will be initialized to the default value regardless of their options.
(c). Avoid using server controls
1. General static display information, try not to use server-side control display. Because the server-side control needs to be sent back to the server for execution,
Will reduce the efficiency of program execution, general use <DIV> display can.
If you use a server-side control, you can increase the efficiency by removing the runat= "server".
2. Disables the status view of the server-side controls, some controls do not need to maintain their state, and can set their properties: Enableviewstate=false;
If the entire page control does not need to maintain a status view, you can set the status of the entire page to false:
The code is as follows: <%@ Page enableviewstate= "false"%>
3. Configure in the Web.config file:
asp.net sessionss can be configured in sessionsstate elements in Web.config or machine.config.
The following are examples of settings in Web.config:
<sessionsstate timeout= "Ten" cookieless= "false" mode= "Inproc"/>
(iv). Avoid using the DataGrid
We all know that the DataGrid is powerful. But while powerful, it increases performance overhead. General use of other controls: DataList
or repeater control can be implemented, try not to use a DataGrid.
(v). String manipulation
1. Avoid boxing operation. The operation efficiency of boxing is low.
For example, run two code snippets:

String Test= "";
for (for int i=0;i<10000;i++)
{
Test = test + i;
}
And
String Test= "";
for (for int i=0;i<10000;i++)
{
Test = Test + i.tostring ();
}
The following code snippet is obviously efficient. Since I is integral, the system first converts the I boxing into a string type and then connects. It takes time.
The reader can copy the test on his machine.
2. Using the Stringbulider class
When string concatenation is made: string str = str1 + str2 + ...;
There are generally more than three connections, preferably with StringBuilder instead of the string class. StringBuilder can avoid recreating a string object caused by
Loss of performance.
Commonly used to assemble SQL statements: Stringbulider.
Readers can test them on their own machines.

3. Try to use less:
Try
{}
Catch
{}
Finally
{}
Statement. This statement performs less efficiently.

(vi). Ado. NET use aspect optimization
1. The database connection is turned on and off. Open when you need to connect, and close the connection immediately after you have accessed the database.
For example, or look at two code snippets:

I.
DataSet ds = new DataSet ();
SqlConnection myconnection = new SqlConnection ("Server=localhost; Uid=sa; pwd=; Database=northwind ");
SqlCommand mycommand = new SqlCommand (strsql,myconnection);
SqlDataAdapter Myadapter=new SqlDataAdapter (QUERYSTR,CONNECTIONSTR);
Myconnection.open (); Open connection
for (int i=0;i<1000;i++)//for loops simulate business logic operations before data is obtained
{
Thread.Sleep (1000);
}
Myadapter.fill (DS);
for (int i=0;i<1000;i++)//for loops simulate business logic operations after data acquisition
{
Thread.Sleep (1000);
}
Myconnection.close (); Close connection
Ii.
DataSet ds = new DataSet ();
SqlConnection myconnection = new SqlConnection ("Server=localhost; Uid=sa; pwd=; Database=northwind ");
SqlCommand mycommand = new SqlCommand (strsql,myconnection);
SqlDataAdapter Myadapter=new SqlDataAdapter (QUERYSTR,CONNECTIONSTR);
for (int i=0;i<1000;i++)//for loops simulate business logic operations before data is obtained
{
Thread.Sleep (1000);
}
Myconnection.open (); Open connection
Myadapter.fill (DS);
Myconnection.close (); Close connection
for (int i=0;i<1000;i++)////for loops simulate business logic operations after data acquisition
{
Thread.Sleep (1000);
}
Show II code better than I code, I early in the connection does not put, if the user a lot of words, easy to connect pool full situation. There is a panic phenomenon when serious.
2. Database query
I. Generate SQL statements directly. SQL Server compiles it every time, and there is no significant performance improvement. It's also not safe enough. Easy to be attacked.
Ii. use SQL commands with parameters. This way SQL Server compiles only once, and the compiled commands can be reused for different parameters. Improved performance.
III. use SQL Server stored procedures. Compile one time.  Independent, easy to modify and maintain. A function that can be used to send multiple statements at a time. Reduced network
Flow. It is not necessarily that stored procedures are more efficient than statements, and if business logic is complex, sometimes statements are more efficient than stored procedures.
(vi). Cache optimization
Caching is divided into two types: page caching and API caching.
1. Using page caching and fragment caching
<%@ OutputCache duration= "5" varybyparam= "None"%>
<%@ OutputCache duration=60 varybyparam= "Textbox1,textbox2"%>
Description: Duration is to set the cache expiration time;
Varbyparam is set to vary according to the parameters, all parameters using the same cache when none
When the TextBox1 is set, it is cached separately according to the different values of the TextBox1, and when there are multiple parameters, the cache is combined;
2.API Cache. For use in applications
I. Example of a cache use:
Http://blog.csdn.net/chengking/archive/2005/10/03/494545.aspx
Ii. attention to Page.cache and HttpContext.Current.Cache differences when used:
They refer to the same object, in page, with Page.cache, if used in Global.asax or in their own class: HttpContext.Current.Cache
In some cases, because it is not HttpContext, use Httpruntime.cache.

The wrong place please the reader to criticize correct!