The Myth Of Linux Community Support. 953

Unless you are stipulating that the server's memory space is being shared by the client, then yes, it all has to be loaded into the client's memory. Everything required to run the program has to be in the client's memory. Whether it gets there locally, or is transfered across the network from a remote mount, it still has to be loaded into the clients memory if the client is running the application.

Alternatively, if you are going to setup all the clients so that they all have the appropriate config files and any necessary shared libs that are not typically already available on the client, then what is the point of keeping just the executable on the server and requiring an NFS mount just to run it? Why not just go ahead and put the executable on the client machines to begin with?

Centralized control of application *deployment* is not the same thing as an *Appication Server*.

Yes, that's a different kind of box. It's called an *Application Server*. Another method would be GUI clients on the client box talking to a backend on the server. Another method would be a web-browser interface to a backend on the server.

I haven't miss out clustering, load-sharing, etc. None of that consbreastutes an "Application Server".

I call that Distributed Computing. Not the same thing as an Application Server. In fact, it's as close to the *opposite* of an Application Server as you can get.

Definitions of Application server on the Web:

"A software server that lets thin clients use applications and databases that are managed by the server. The application server handles all the application operations and connections for the clients. securityresponse.symantec.com-avcenter-refa.html"

...

"A server in a client-server network which runs one or more applications that can be shared by client stations and which also shares the data processing burden with client stations. This shared application and shared data processing model contrasts with the model used for other servers, such as file servers, that simply send, receive, and store files, requiring client stations to run all applications and process all data. Either model can be most advantageous, depending on circumstances. In many circumstances the application server model allows for faster data processing, faster throughput to client stations, greater data reliability, and increased data security. www.orangecom.com-glossary.HTML"

...

"A computer that handles all operations between a company's back-end applications or databases and the users' computers' Web browsers. www.ifactor.uk.com-glossary-main.htm"

"a server computer that: Executes applications when requested by other computers. Stores business logic and the business model clbuttes of applications. Serves requests for dynamic HTTP webpages from web servers. www.donald-firesmith.com-Glossary-GlossaryA.html"

*******************************************************************

Definitions of Distributed Computing on the Web:

"Distributed computing is a programming model in which processing occurs in many different places (or nodes) around a network. Processing can occur wherever it makes the most sense, whether that is on a server, Web site, personal computer, handheld device, or other smart device. www.microsoft.com-net-basics-glossary.asp

" a type of system that divides a workload to computers connected to a network. The network may be either be enclosed in a room or out in the open, like the Internet. Distributed computing is also referred to as distributed processing, cooperative computing, and collective computing. See the Introduction section for a major treatment. See also: render farms, parallel processing. Pictures: Regular Computing vs. Distributed Computing diagram. library.thinkquest.org-C007645-english-glossary-start.htm"