Creating Custom Cache Dependency

ASP.NET 2.0 offers you several ways to set a dependency between a cached item and a file(s), another cached item(s) or SQL Server database table. No doubt they satisfy most of the real world needs. However, at times the features offered by these dependencies are not sufficient. In such cases you can create your own dependency and use it instead of inbuilt ones.

Coding style

The program should be easy to read and to understand when you need to refer back to

it. Follow these guidelines while coding:

§ Always use "Option Explicit" to catch any undeclared or misspelled

variables. Also, the use of "Option Explicit" makes the Web pages run

fast. The "Option Explicit" option forces the explicit declaration of

variables.

§ Declare one variable per line. This avoids confusion about datatypes.

§ Use comments wherever possible to document a difficult code section.

§ Use blank lines in the code for clarity.

§ Use proper code block indenting.

Programmability

The ADO.NET model uses typed programming to manipulate objects. In typed

programming, the programming environment or programming language itself recognizes

the types of things that are important to users. To take full advantage of typed

programming, you must know the things that are of interest to programmers and to end

users. Consider the following code using typed programming in ADO.NET:

If TotalQty > DataSet1.ProductInfo("Baby Food").QtyAvailable

This code is equivalent to a line using non-typed programming and is easier to read by

end users. An end user who has little or no programming experience can easily grasp

the meaning of the condition being tested. Also, in non-typed programming, if the

developer makes a spelling mistake by chance (for example, ProductInfo is spelled as

ProdcutInfo), a run-time error will get generated. On the other hand, in typed datasets,

errors in the syntax caused by misspellings are detected at compile time rather than at

run time.

Maintainability

After an application is deployed, there might be a need for changes in the application.

For example, the application might need substantial architectural changes to improve its

performance. As the performance load on a deployed application server grows, system

resources can become inadequate, resulting in higher response times. As a solution to

these problems, the application might need to undergo architectural changes by adding

tiers. Here, the problem is not the multitier application design, but rather the problem lies

in increasing the number of tiers after an application is deployed. This transformation

becomes easier if the original application is implemented in ADO.NET using datasets. In

ADO.NET, the communication between tiers is relatively easy, because the tiers can

transmit data through XML-formatted datasets.

Interoperability

The ADO.NET model is designed to take maximum advantage of the flexibility provided

by the large industry acceptance of XML. ADO.NET uses XML for transmitting datasets

among components and across tiers. Any component that is capable of reading the XML

format can process the data. It is not necessary for the receiving component to be an

ADO.NET component. The component that is sending or transmitting the dataset can

simply transmit the dataset to its destination without bothering with how the receiving

component is implemented. The component asking for the dataset, the destination

component, can be implemented as a Visual Studio application or any other application.

However, the important point to be considered is that the receiving component should be

capable of accepting the XML file formatted as a dataset.

Scalability

The Web-based, data-centric applications require multiple users to access data

simultaneously. This increases the demand on data to be accessed, making scalability

one of the most critical features. Applications that use resources, such as database

connections and database locks, cannot support more users to access data

simultaneously, because eventually the user demand for the limited resources will

exceed their supply. Because ADO.NET uses disconnected data access, applications do

not retain database locks or active database connections for long durations. Hence,

ADO.NET accommodates scalability by encouraging programmers to conserve limited

resources, and allows more users to access data simultaneously.

ADO.NET Object Model

The .NET Framework is designed to change dramatically the developer's current style of

developing applications, including the data access features. For the .NET applications,

the primary data access technology to be used would be ADO.NET — the latest addition

to the ADO model.

The ADO.NET Object Model is primarily divided into two levels:

§ Connected Layer: Consists of the classes that comprise the Managed

Providers

§ Disconnected Layer: Is rooted in the DataSet

This section describes both the Managed Providers and the DataSet.