Summary

In this chapter, we have focused on different clustering techniques and some of the benefits and concerns of these.

Clustering is used mainly to solve two things:

• Scalability: Being able to handle an increasing load

• Availability: Being available for as much of the time as possible

You have seen that Microsoft offers two techniques to help you with this. The first, Microsoft Cluster Service, is great for enhancing availability. The second, Network Load Balancing, is great for improving both scalability and availability.

But with clustering several servers together arises the problem of managing them. Another major issue is updating, for example, a large Web cluster, so that all member servers receive the changed files when new versions are available. To help you with this, Microsoft provides the Application Center server. In this server are tools and wizards that ease the administrative issues involved in having large clusters. Application Center also makes it easy to administer several clusters, local as well as remote, by being a single point of control.

Another great tool in Application Center is Component Load Balancing, which offers the possibility to distribute COM+ activation requests just like you distribute HTTP requests with Network Load Balancing.

You have also seen that using session state in a load-balanced cluster decreases performance and should not be used if you want a scalable solution with optimized performance.

Throughout this book, you will find more tips for creating a scalable .NET solution.