In the days of always-available Web content, it's a surprise that many people I talk to do not know how to achieve a high level of uptime. There are many tools that can help. One of them, Application Center 2000, is Microsoft's deployment and management tool for high-availability Web applications built on the Microsoft Windows 2000 operating system. Application Center 2000 makes managing groups of servers as simple as managing a single computer. In this article, we will take a closer look at the product.
Clustering, load balancing and high availability
Before we examine the actual product, let's consider why it was created and where its overall usefulness may be of help to you and your organization. Clustering and load balancing solutions offer you the ability to have some type of fault tolerance and disaster recovery. A clustered solution (we will use the example of two servers for now to keep it simple) will allow you to have some recovery functionality over your systems -- if one server crashed, your clients would still have access to your company's data. On the other hand, a load-balanced solution will allow your network clients to have more available access to company Web sites or data. Application Center 2000 works with the load-balancing portion of clustering. Load balancing with Microsoft Windows 2000 Advanced Server is doable, but is definitely not very manageable. You can set up network load balanced (NLB) solutions without Application Center 2000, but with it, you have much more to gain (as we will see shortly).
Application Center 2000 basics
Application Center 2000 is a software package in the .NET Enterprise Server family. It installs on top of a server running either Windows 2000 Server or Windows 2000 Advanced Server. You can also use the product with Datacenter. Application Center 2000 allows you to manage all your NLB nodes under one console, something that was not previously possible. (Again, do not confuse this with Windows 2000 Clustering Services, which does allow you to have a centralized console by default -- we are discussing the NLB portion of high availability here.) Once you install the Application Center 2000 product, you are able to manage up to 32 nodes in one console. The console uses the standardized Microsoft Management Console (MMC) snap-in format as seen below.
|Application Center 2000 console|
You are now able to manage all nodes under one console. You can also install the console on any workstation to manage all nodes from another location.
Application Center 2000 features
Application Center 2000 offers several features that provide application management, software scaling, and mission-critical availability.
- Simplified application management -- unless you have the need for a feature such as this, Application Center 2000 may not be a necessary tool for you. If you do have the need, then application management is one of the biggest selling features of this product. If you are a Web site administrator and you have content spread across a two-node NLB cluster, you have no way to optimize the deployment of content to your servers unless you use a product like Application Center 2000. It gives you plenty of ways to roll out content to your sites, including pages like *asp, and *.htm, and you can also roll out actual COM+ components and set up NLB or clusters for component load balancing.
- The manageability console is another big selling point. The Application Center 2000 MMC allows you the snap-in to manage all servers at once, as one group. You can apply a change to one server in the cluster (usually the cluster controller) and apply the content to all other servers in the group at the same time. This makes application updates easier to manage.
- When you want to actually deploy content to a site (like a Web e-commerce site), you may want to make sure such tasks are simplified. Simplifying the task of migrating applications through the development cycle (from development to testing to production) is a major undertaking and Application Center 2000 aids you in this cycle. Application Center 2000 helps the web administrator to automate the deployment of applications from one server to the other servers, releasing better content (and more stable content) to your nodes in the load-balanced cluster.
- Scaling features are simplified. You can have a node act as a "cluster controller" and you are able to add nodes, remove nodes and change the status of the cluster controller to aid in repairs and scaling. Scaling is either up or out. Scaling up is building a single node through upgrades to hardware. Scaling out is adding more nodes. You have tremendous scaling-out potential with Application Center 2000.
- Other scaling features will allow you to scale out to 32 nodes when needed. You company may find the need to add more load balancing nodes as traffic increases either over time or during specific times like holidays and during peak business periods. You can add or remove nodes at will when needed.
- You have automation features with Application Center 2000 that are more advanced than what you could do without it. Specific events of health can be monitored and, when triggered, can automatically send an e-mail or ring a pager number. This is crucial to using actual fault-management capabilities.
- Application Center 2000 gives you advanced monitoring of performance with "Health Monitor," which does advanced monitoring of separate nodes on the cluster and also gives you performance monitoring of the entire cluster. That is needed when you have to report baselines of the entire solution, not just one piece of it.
In sum, think about implementing this solution if you are in the business of deploying large Web farms or Web content, or if you need to heavily manage and monitor performance on NLB clusters. Application Center 2000 has much to offer in this area, so download the trial software and see if it fits in your organization.
For more information:
Application Center 2000 overview
Application Center 2000 architecture
Application Center 2000 monitoring
Application Center 2000 trial software
Application Center 2000 downloads
Application Center 2000 downloadable resource kit
Application Center 2000 product guide
About the author:
Robert J. Shimonski (Truesecure TICSA, Cisco CCDP, CCNP, Nortel NNCSS, Microsoft MCSE, MCP+I, Novell Master CNE, CIP, CIBS, IWA CWP, Prosoft CIWSA, SANS GSEC, GCIH, CompTIA Server+, Network+, Inet+, A+, e-Biz+, Symantec SPS and NAI Sniffer SCP) is a lead network and security engineer for a leading manufacturer company and a part time trainer. Robert's specialties include network infrastructure design, network security design and management, systems engineering and troubleshooting with Sniffer-based technologies. Robert is the author of many security related articles and published books, including the upcoming "Sniffer Network Optimization and Troubleshooting Handbook" from Syngress Media Inc (ISBN: 1931836574). You can contact him in the Net Management Forum, or send him an e-mail at firstname.lastname@example.org.
This was first published in July 2002