bearkat
asked on
LAN and subnetting
I have inheirited a network with 140 workstations and 8 servers. They are all setup using one subnet 10.1.0.0 255.255.0.0 so they are all on one subnet and on vlan1. I captured some traffic using wire shark and the traffic hitting a workstation is on average 4450 bytes/sec. No one is complaining of slow performance but I was wondering if this network could benefit from benig subnetted and if the bytes per sec was an acceptable number. Thanks for your opions in advance.
4450 bytes/sec = 35.6Mbps. If you're on a 100Mbps network that's not unreasonable.
Would you benefit from going gigabit? Maybe. Is it worth the investment? Probably not if there are no other issues.
Would you benefit from going gigabit? Maybe. Is it worth the investment? Probably not if there are no other issues.
4450 bytes/sec = 35.6Mbps
Wait.... what?
4450 bytes/s = 4.45 kbytes/s = 35.6 KILObits/s = 0.035 MEGAbits/s
Wait.... what?
4450 bytes/s = 4.45 kbytes/s = 35.6 KILObits/s = 0.035 MEGAbits/s
Also, if he had 35.6 Mbps broadcast traffic, that would be VERY VERY BAD.
Fair enough - the OP wrote 4,450Bps and I read 4,450kBps.
The OP may want to clarify, because it's unlikely they're running at modem speeds and no one's complaining.
The OP may want to clarify, because it's unlikely they're running at modem speeds and no one's complaining.
ASKER
All of our switches on campus are gigabyte switches running in full duplex. I dont see any problems but just wanted to double check with the great minds on EE.
One workstation doesn't likely tell you what's going on. A better place to look is on a busy server or on the internet gateway - which you might access on a managed switch mirrored port and your own laptop or workstation.
One measure is "is anyone complaining". That's a rather compelling measure.
Another measure is "will anyone be complaining soon?". That's tougher to determine and will be fraught with technical opinions, measurements, etc. etc.
A hybrid might be: "we will have plenty of time to make changes if and when there start to be reasonable complaints". This also preserves capital and is based on a degree of technical confidence. These things don't happen over night.
You didn't say what the internet up/down speeds are.....
One measure is "is anyone complaining". That's a rather compelling measure.
Another measure is "will anyone be complaining soon?". That's tougher to determine and will be fraught with technical opinions, measurements, etc. etc.
A hybrid might be: "we will have plenty of time to make changes if and when there start to be reasonable complaints". This also preserves capital and is based on a degree of technical confidence. These things don't happen over night.
You didn't say what the internet up/down speeds are.....
ASKER
Internet speeds are 30mb down 4 mb up and that I have looked at on multipule workstations. I have a Barracuda link balancer with a T1 and cable connection. No one has complained but I want to be ready ahead of time so I dont have to hear it. lol... What are some good test I can do to check on network speeds locally and to the internet?
Two good ones for Internet speed tests. Note these will be bound by upstream traffic, so don't rely on them as a true measure of your maximum bandwidth.
http://www.speakeasy.net/speedtest/
http://www.dslreports.com/stest
http://www.speakeasy.net/speedtest/
http://www.dslreports.com/stest
ASKER CERTIFIED SOLUTION
membership
Create a free account to see this answer
Signing up is free and takes 30 seconds. No credit card required.
ASKER
The Barcuda link balancer keeps stats on bandwith usage and we never go above 30% of our available bandwith. Well after everything I have heard here what I thought was right. It works and works well on a flat subnet so there is no reason to mess around with it at this point. Thanks for the help guys
If everything is working, DO NOT fix what is not broken :)