Architecture Decision - Agent versus Agent Less

J

Jenbo

This is a very general question I am asking as I am at the start of a
design and analysis phase of a project and I am still thinking of
exactly what way to architect the system I am constructing. Basically I
will have an engine system centralised on client's server, this will be
SQL with a management web app on front of it. This needs to communicate
with the client's pcs, however many number of them there are. Based on
this communication an action will be taken on the client's pc that will
run an app. So, this is fine to do in a windows app, make it a system
tray icon based guy and it can launch apps, web sites, do whatever, and
it can communicate with our engine as well, over whatever mechanism.
This would be an agent based system say. Now, say I had a client who
said that we couldn't install any software over their network of pc's,
what would be the way to do this? Have some kind of group policy in
Active Directory that launched the application from a network location
and then did it's business that way. This is what I would call agent
less. Would this have more network over-head than the agent based
system, as it would have to network to the pc and back to the engine
also to get instructions? I know this is vague but any advice would be
welcome.

Cheers

J
 
M

Michael Nemtsev

Hello Jenbo,

J> This is a very general question I am asking as I am at the start of a
J> design and analysis phase of a project and I am still thinking of
J> exactly what way to architect the system I am constructing. Basically
J> I will have an engine system centralised on client's server, this
J> will be SQL with a management web app on front of it. This needs to
J> communicate with the client's pcs, however many number of them there
J> are. Based on this communication an action will be taken on the
J> client's pc that will run an app. So, this is fine to do in a windows
J> app, make it a system tray icon based guy and it can launch apps, web
J> sites, do whatever, and it can communicate with our engine as well,
J> over whatever mechanism. This would be an agent based system say.
J> Now, say I had a client who said that we couldn't install any
J> software over their network of pc's, what would be the way to do
J> this?

You can ship the package for the local deployment. But how is going to
communicate with your server? Does he want the intranet version of your product?

---
WBR,
Michael Nemtsev [C# MVP] :: blog: http://spaces.live.com/laflour

"At times one remains faithful to a cause only because its opponents do not
cease to be insipid." (c) Friedrich Nietzsche
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top