You are currently on IBM Systems Media’s archival website. Click here to view our new website.


Bookmark and Share
RSS

Recent Posts

N-Tier Architectures

November 21, 2013

Editor’s note: This is the fourth blog entry in a series.

In my previous post, I wrote about remote procedure call, remote database access, distributed transaction processing and message queuing. All are client-server (CS) functions with mainframe origins. In this post, I want to discuss the distributed architectures they were designed to support.

Centralized computing was considered a one-tier architecture. The term “one-tier” really doesn’t capture the diversity and flexibility of the integrated model that contained multiple LPARs, multiple address spaces and robust systems like CICS and IMS that handled task dispatching and provided many application services.

The two-tier model consists of the client software at the workstation and a server-tier that runs database, application and network (Web) software. Having most of the functions on one server has some limitations depending on the robustness of the hardware and software but it is an easy way to exploit off the shelf low-cost components. Also, you can get a departmental-type solution up and running quickly.

The three-tier model uses a two-server implementation where the database engine is on its own server. In some cases, this improves the robustness of the solution and makes it easier for the database engine to serve multiple application servers. Depending on the underlying hardware, three-tier solutions scale better than two-tier implementations as long as the networking connection is reliable and consistent.

Today, we live in an n-tier world. Depending on the design of the application, we often find Web, application and database software running on different servers in different address spaces. As we have flexible interfaces available from mainframe software, central computers now participate fully in many different roles in these n-tier constructs. The ultimate n-tier solution is an enterprise-level one that makes use of corporate data gathered and processed wherever it is found in the company.

Do you work in an n-tier environment? In the next post, I will compare early middleware to the modern middleware of today.

 

Posted November 21, 2013| Permalink

-->