Yep. I agree with this completely.
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Joe Pluta
Sent: Friday, November 09, 2007 9:38 AM
To: 'Midrange Systems Technical Discussion'
Subject: RE: Surveillance Videos and Security Cameras on the System i ? ? ?
? ?
From: Shannon ODonnell
What is the benefit of streaming that data to the iSeries if you have to
have the PC on the front end to interface with the video hardware and the
software codecs and to view the images?
Shannon, I think the point folks are trying to make is that the hardware
used for the client has nothing to do with the hardware used for the server.
And in that, they're probably right, since in the best solution you probably
want remote access to your data, which would be easiest through a browser
interface.
The point I guess is that if your server is simply streaming data out a
TCP/IP socket to a client, the server can be a System i as easily as a
Windows or *nix box, with all the standard benefits of the System i (backup
and recovery and security and auditing and so no).
My issue with the concept is simply the amount of data being streamed out.
Inbound data load is pretty easy to calculate and configure for, but if
you're doing a lot of outbound, you might have issues. If you're simply
doing an occasional retrieval of stored data, that's one thing. If,
however, you need to send the streaming web data live to multiple monitoring
stations, then suddenly you've got a huge jump in bandwidth.
My opinion would be that a System i would work best with a two-tiered
system, one with live streaming capability to monitoring stations and then
archive retrieval from a server. The System i would be the archival server
and would simply be one of the live feeds. It would receive the live feed
and store it.
Joe
midrange.com
