SG Broadband Community

ok.. Here we go...

I am one of the IT guys here for a small 24 user network..


We are running this..

Dell Dual 1.2 GHZ with 2 gig ram Server with Win2K
Linksys 24 port 10/100 Switch
24 Windows XP Professional Work Stations with 100Mps connection to switch..

Internet Connection: Fractional T1 256Kbs (slow i know)

here is my argument that i am having with our so called IT manager...

We run a DOS based inventory software on the network.. It is programmed in FOXpro and uses DFM files. (again another dinosaur here) This software does not use the internet at all..

-Argument-

He is stating that because i listen to steaming music over the internet, that is slows the inventory software down.

I told him that he was on crack, because the 5K of bandwidth that is used to listen to music would not effect the 100M of connection we have from the server to the workstation for running the inventory software... I tried to tell him that the reason the software slows down is that it is DBF files and the more users that are signed on and doing things, the slower the software will run..

His reply was that he thinks it still slows it down...

It is prob that the server runs win2k and hates this dos based crap that we use...

Can anyone help me in convincing this guy to learn the definition of bandwidth or am i totally way off??

thanks...

skeeter...

I am no IT Pro, but it seems to me that a real simple solution would be to run the software without the streaming music running, and see what happens. If the software runs the same, then you are in the clear.

Like I said, "seems" like a real simple solution.

Bandwidth probably wouldn't be the bottleneck - more CPU. Decoding anything takes CPU cycles. Set up performance monitoring on the server (% processor time per process, private bytes per process, and network throughput). Set a baseline when no one is on the system at all, then in normal operation (with your DOS app) without music. Finally try it with your music. You would have three sets of data and could look at the difference between them. I think you can even save them as a pretty graph or pie chart Managers like pictures...they take them to that happy place.


Skye

LOL, the happy place.. I've been there before.

You may be forgeting frequency of users. If he started the monitoring process and there were 5 users logged on using resources; then you start the second monitoring process without streaming music and an additional 10 users logon while using more resources, you will get unfair results. If you have a late shift, try at that time when you have a minimal amount of users on so it will not impact the monitoring results as much. The basic is that bandwidth is being used by all and you are still using bandwidth. Is it slowing down your processes, I don't know but it will be very minimal impact to the network users. I would still run the monitoring and show that there was next to no difference.

I think cyberskye is on the right track. Untill you do some analyzing of the network/server, you don't have a clue to what's going on.

For one I wouldn't put a Linksys switch on a bandwidth intensive network. They just don't perform that well when every port starts getting hit hard. Also some accounting and inventory software is very network intensive in the fact that it's not transfering large files, but many files often. I think in this situation streaming music can cause collisions, which takes up bandwidth because of retransmissions of the data.

Try running Anasil and see what your network is doing. Just because a network is rated for 100Mps in theory, doesn't mean it will get anywhere near that. Things such as cable/connection quality, uneeded protocols such as IPX/SPX and netBEUI will slow things down (network chatter), along with networked printers.