Show Posts
1
Users / Question about thin clients.
« on: December 24, 2007, 10:28:20 pm »
First, some definitions:
Are LinuxMCE thin clients true thin clients, or are they just diskless nodes?
On a diskless node, apps all run locally. The filesystem may be remote, but all the processing is still local.
On a real thin client, *everything* is on the remote computer. On a real thin client (such as LTSP on GNU/Linux, or those weird SunRays on Solaris), absolutely everything, including processing and so on, runs on the server - it's a bit like VNCing into the server actually, except not so crude or bandwidth-hungry.
So, are LinuxMCE thin clients fully thin, or are they merely diskless nodes?
Supposing that they're diskless nodes, would it be possible to install LTSP on the Core machine and therefore have true thin clients?
TRUE THIN CLIENT
A machine that boots an OS image off the network, then uses that OS to log into a central server.
DISKLESS NODE (a.k.a., FALSE thin client)
A machine that boots an OS image off the network, then uses that OS as a normal fat-client OS.
Are LinuxMCE thin clients true thin clients, or are they just diskless nodes?
On a diskless node, apps all run locally. The filesystem may be remote, but all the processing is still local.
On a real thin client, *everything* is on the remote computer. On a real thin client (such as LTSP on GNU/Linux, or those weird SunRays on Solaris), absolutely everything, including processing and so on, runs on the server - it's a bit like VNCing into the server actually, except not so crude or bandwidth-hungry.
So, are LinuxMCE thin clients fully thin, or are they merely diskless nodes?
Supposing that they're diskless nodes, would it be possible to install LTSP on the Core machine and therefore have true thin clients?