ASP.NET Web Cache Spurs Performance and Scalability

By Iqbal Khan

Applications based on ASP.NET, Microsoft's Web-application framework, are making greater inroads into the enterprise. At the same time, bottlenecks resulting from growing numbers of users and transactions continue to prompt IT professional to call for improved performance and scalability.

The Problem: ASP.NET Bottlenecks

Bottlenecks can occur in ASP.NET applications for a variety of reasons. The most obvious: Data-storage technology isn't as scalable as Web-applications architecture. Any place in an ASP.NET application that deals with data storage or data access immediately becomes a logjam when you try to scale your application. Two areas where this happens are Session State storage and application data from a relational or mainframe database.

Another bottleneck occurs if your ASP.NET application is making service- oriented architecture (SOA) calls to Web services. Here, the slowdown happens because the Web services have the same issues as your ASP.NET application (namely, in data storage and access). Chances are that a Web-services farm is being shared across multiple applications and, therefore, being stressed much more than any one ASP.NET application, creating the scalability bottleneck.

Bottlenecks can also occur between the user's browser and the ASP.NET Web farm. These clogs are related to the fact that ASP.NET pages have to be executed repeatedly at time involving intensive CPU processing. This process also involves sending heavy data elements (images, documents, etc.) to the user again and again.

In a previous TechNet Magazine article, I discussed ASP.NET performance and scalability issues, focusing on Session State and application data (see "Providing Scalability for ASP.NET Applications," June 2009). In that article, I covered how those problems occur, including the reasons that the ASP.NET Session State becomes a logjam as a Web farm grows. I discussed the fact that distributed in-memory cache is a superior alternative to Microsoft's existing storage option for ASP.NET Session State. I described how application data coming from a database can cause scalability bottlenecks. I also detailed how distributed caching resolves those ASP.NET Session State storage bottlenecks with the help of different caching topologies that each offer different features, but all address scalability and assure 100 percent uptime.