Client Server - Myths and definitions

Forum related to the ZDBC API-Layer

Moderators: gto, cipto_kh, EgonHugeist

mparak
Senior Boarder
Senior Boarder
Posts: 81
Joined: 12.09.2005, 06:51
Location: Durban South Africa
Contact:

Client Server - Myths and definitions

Postby mparak » 13.07.2006, 16:04

Hi folks,

I was reading up on Michael Seeger's (ZeosLib QA- and Documentation Team) explaination on the use of TZtable component and I quote:...
--------------------------------------------------------------------------------
TZTable

TZTable acts like BDE's TTable. As a principle you only should use TZTable in a C/S application if you have very small tables because all records of the table will be transferred from server into client's memory by opening the TZTable. This is a bahaviour similar to a "SELECT * FROM XYZ" statement. You should even prevent a statement like this in a C/S application. The intension is to keep the resultset that has to be transferred from server to client as small as possible (perferably onle one record).

Michael Seeger - ZeosLib QA- and Documentation Team
--------------------------------------------------------------------------------
I then doubted my basic definitions of Client server and came back with this definition:...........

Client-server
From Wikipedia, the free encyclopedia

Client-server is a network architecture which separates the client (often a graphical user interface) from the server. Each instance of the client software can send requests to a server or application server. There are many different types of servers; some examples include: a file server, terminal server, or mail server. While their purpose varies somewhat, the basic architecture remains the same.

Although this idea is applied in a variety of ways, on many different kinds of applications, the easiest example to visualize is the current use of web pages on the internet. For example, if you are reading this article on Wikipedia, your computer and web browser would be considered a client, and the computers, databases, and applications that make up Wikipedia would be considered the server. When your web browser requests a particular article from Wikipedia, the Wikipedia server finds all of the information required to display the article in the Wikipedia database, assembles it into a web page, and sends it back to your web browser for you to look at.

----------------------------------------------------------------
Is it only me or has the author Michael Seeger gotten the entire concept, backwards. If the Ttable issues a select * from Xyz then all good programmers know that it should be used to open LOCAL tables and not Client server ones over the wire. I would not advise the use of tzTable in the C/S environment.

What do you think...

Regards

M.

User avatar
mdaems
Zeos Project Manager
Zeos Project Manager
Posts: 2766
Joined: 20.09.2005, 15:28
Location: Brussels, Belgium
Contact:

Re: Client Server - Myths and definitions

Postby mdaems » 13.07.2006, 22:23

As a principle you only should use TZTable in a C/S application if you have very small tables because all records of the table will be transferred from server into client's memory by opening the TZTable. This is a bahaviour similar to a "SELECT * FROM XYZ" statement. You should even prevent a statement like this in a C/S application. The intension is to keep the resultset that has to be transferred from server to client as small as possible (perferably onle one record).

Michael Seeger - ZeosLib QA- and Documentation Team


Well M.,

I think it's you. Doesn't Michael say explicitly you should prevent (avoid) using these TZTable in CS? And ifyou do,just use itfor very small tables?

Mark :P :wink:

mparak
Senior Boarder
Senior Boarder
Posts: 81
Joined: 12.09.2005, 06:51
Location: Durban South Africa
Contact:

Postby mparak » 15.07.2006, 22:41

I obviously read him wrong.
My Bad


Return to “ZDBC (ZEOS Database Connectivity)”

Who is online

Users browsing this forum: No registered users and 1 guest