Computing architecture

Details Computing architecture Computing architecture

2000-01-14

2003-11-25

Jaroenchonwanit, Bunjob (Department: 2141)

Electrical computers and digital processing systems: multicomput

Distributed data processing

Client/server

C715S252000

Reexamination Certificate

active

06654784

ABSTRACT:

Appendix A containing a computer program listing is submitted on two identical compact disks. Each compact disk contains a file entitled Appendix A.doc. The file is a Microsoft Word document created on Apr. 8, 2003 and is 255 KB in size. Appendix A is hereby incorporated by reference.
FIELD OF THE INVENTION
This invention relates to computing architecture, and more particularly to a method and apparatus for running applications solely on a backend server, with the Graphical User Interface of the application running on the client machine in the form of HTML-based web pages.
BACKGROUND OF THE INVENTION
Computing, in general, can be classified into two categories: Personal Computing and Enterprise Computing. Enterprise Computing, is concerned with the computing activities of an organization and the information infrastructure, including hardware, software and services to support these activities. An organization's computing activity includes not only that associated with each individual within the organization, but also the collaboration among different members or groups which requires access of different members to different computing resources. Also involved is the collaboration and interaction with other related organizations. In short, Enterprise Computing relates to the information infrastructure to provide each individual entity with access to the right application and data at the right time and the right place.
Traditionally, the architecture of Enterprise Computing is client/server computing, especially Fat Client Computing. Fat Client Computing refers to a computing model in which applications run on both client and server machines, either stored locally or downloaded. Data is also stored locally. A high bandwidth network is used to provide communication between the server and client. In short, Fat Client Computing requires that both the client and server possess enough computing power to run an application and a high bandwidth communication mechanism.
Today, Fat Client Computing is ubiquitous, typically involving server machines such as the Unix/WinNT server, client machines such as PCs running Windows OS, and connectivity being either a high speed LAN or WAN. Though it offers flexibility and lower initial acquisition cost than its predecessors, there are problems associated with Fat Client Computing.
The main concern is the total cost of ownership. Currently the following are four main problems associated with enterprise computing due to its Fat Client Computing architecture:
From a management perspective, traditional enterprise-wide application deployment is often time-consuming, expensive and difficult to maintain. Not only do administrators have to physically distribute applications to every client machine, they also have to deal with version control issues, remote support, multiple system configurations, data replication and protection. When confronted with thousands of users, the cost of application ownership is quickly out of control.
Today's corporate IT system comprises of a mix of heterogeneous hardware systems and platforms, network connectivity and operating systems. Many applications run on only a certain platform and are installed on only certain computer systems that people from other platforms or other computer systems have problems in accessing. Workers spreading around the world want to access the same set of data and information such as employee information. Companies want to let customers and suppliers access some information traditionally only available within the company, such as order status, but with no idea of what kind of computer system their customers and suppliers are using.
Moreover, most corporate applications today are designed for high speed networks and powerful computer systems. This type of design puts tremendous strain on corporate networks, and the growth of the network bandwidth cannot keep up with demand. Additionally, this type of design yields poor performance on lower bandwidth networks, or is even impossible to run due to bandwidth limitations. Moreover, much of the time, a slow network such as the Internet is the only available connection. Because of this, many users simply refuse to use vital applications and access vital data to get their work done. When this happens, redundant work and significant decreases in productivity may incur.
Additionally, security is a big challenge. In the current enterprise computing architecture, critical applications and data reside both on clients and servers. They are distributed over the whole enterprise. Not only does this increase the risk of unauthorized access to critical information, but also the distributed nature of the applications and data increases the risk of loss or stolen information.
Due to the problems of Fat Client Computing, another computing architecture, called Thin Client Computing, is gaining momentum. Thin Client Computing is a generic term used to describe a group of rapidly emerging technologies that attempt to solve the problems of Fat Client Computing and provide a reduction in total cost though a combination of reduced hardware cost, reduced maintenance and support cost, reduced LAN/WAN bandwidth requirements, reduced down time, improved performance and enhanced security.
Thin Clients bear a resemblance to the dumb terminals in the mainframe age. They rely on the server in order to operate. Most, or, 100%, of the applications run on the server, while the Thin Client acts only as a user interface.
Currently the most widely accepted Thin Client Computing model is based on X-Windows, Microsoft Terminal Server and Citrix technologies. This computing model completely separates an application from its user interface. While the application is 100% running on the server, its user interface is presented on the client machine, which is purely used for user interaction. The user's input, such as keyboard strokes/mouse movements, originate at the client machine and are transmitted back to the server via the network. The server's application processes these events and sends its output to the client machine. The communication between client and server is through well-established protocols.
A protocol is a mutually agreed way to present display information and user input information. If the server is a Unix or Linux machine, the protocol will be the X-Windows Protocol. If the server is a Windows machine, the protocol will be the Microsoft Remote Display Protocol, RDP, or Citrix Independent Computing Architecture, ICA. Currently there are three kinds of clients: ie. a native client, a Java client and a web-client.
A native client means that the client program is a native program running on the client machine's operating system. If the client machine uses a Windows system, the client program has to be a Windows program. A Java client means that the client program is written in Java. By leveraging Java's “Write Once Run Anywhere” capability, a Java client is not limited to a particular kind of Operating System. A web client means that the client program is running inside a web browser, and is either a Java Applet/ActiveX or a browser plugin.
Thin Client Computing in general solves the problems associated with Fat Client Computing by compressing bandwidth and making the client program much smaller. Moreover, Thin Client systems rely on Java so that users do not need to install the program. One simply goes to the web browser and launches the program using Java applets. This implies that applications and data can be centrally managed and maintained.
Nevertheless, Thin Client Computing also has its own set of problems.
First, Thin Client Computing still requires a reasonable network bandwidth, which is higher than the current Internet. Most of the time, the user still needs to install a client program on his/her computer, which is exactly what Thin Client Computing trying to avoid, though the program is much smaller than Fat Client programs. In some cases, the user may not need to install a client program on his/her computer by running a Java Applet or Act

Affiliated with

Also associated with

Rating

Computing architecture does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Computing architecture, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Computing architecture will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3164837