Slow and fluctuating server network speeds in a new data center

Posted on 2010-09-15
Last Modified: 2012-05-10
We have an FTP server in a data center and we just switched to a new network provider.

Before the switch, things were just fine.

We then switched to a new network provider in the exact same building (there are multiple companies that provide network services in that data center) and things have been very troublesome...

The speeds kept fluctuating and were at first extremely low, like fluctuating between 20 kBps and 100 kBps. Usually more like 20 kBps.

THEN- They said they changed the port speed on their router to auto-negotiate, and that improved things a lot, but we still have serious connection issues.

The speeds are now about 100-200 kBps and some people have no problems at all, but some people have a lot of trouble with proper speed and, while its unclear whose end it is on, even some supposed dropped connections.

Our VMware server, which is on the new network, is also having trouble with dropped connections.

Another thing is, maybe its just me, but the CPU seems busier than usual.
Question by:the_cyman
  • 5
  • 4
LVL 57

Expert Comment

ID: 33691107
Whomever controls the switch that their router connects to needs to work with them and get the port speed and duplex the same.  It sounds like a duplex mis-match.  

Even with both sides set to auto, they may not negotiate correctly.  Unless you have gigabit fiber on both ends, hard coded the maximum speed and duplex is best.

What is the bandwidth you have?  If you are getting 200 kBps max, that is just a bit over T1 (1.54 Mbps) speeds.

Do you know of other companies in the same data center that are using this same provider?  Are they having problems?

Author Comment

ID: 33692976
No, I don't think others are having issues but its hard to tell. We currently use a unmanaged switch. Is there any way to force the switch to 100 Mbit?
Or will we need to get a new switch?

Author Comment

ID: 33693098
Also, is it possible that our network bandwidth just stinks? Like, maybe the VMware issue is separate. But as for others, maybe the transfer speeds are just poor because the provider is poor. Maybe they use Cogent... But to get the magnitude of speed difference with this new provider:
- Someone in India who got 90 kBytesps now gets 40 kBytesps
- Someone who gets 1000 kBytes ps now gets 200-300 kBytes ps

All the above are relatively far from us.

Also... Our upload speed is like WAY faster than our download speed. We are not very congested, maybe 10 simultaneous transfers at most.

Our connection is fiber, 100 mbits/ps
Windows Server 2016: All you need to know

Learn about Hyper-V features that increase functionality and usability of Microsoft Windows Server 2016. Also, throughout this eBook, you’ll find some basic PowerShell examples that will help you leverage the scripts in your environments!

LVL 57

Expert Comment

ID: 33693982
If you truly have 100 Mbps Internet connection then more than likely any bandwidth issues should end up on the remote site as 100 Mbps is typically way more bandwidth than what most people have.  Unless you have hundreds or thousands of connections tring to transfers at the same time.

You may want to go to and pick a test server near you and see what you get.

Author Comment

ID: 33694532
Ok, so how do we change our port speed if we have an unmanaged gigabit port?

The connection setup is this:
PROVIDER'S CISCO ROUTER -> Our DLink DGS 1024D (unmanaged switch, in our colo rack) -> Our 3 servers.

Would we need to get a managed switch in this case?
LVL 57

Expert Comment

ID: 33694768
Did you do the speed test?  What did you get?

If you wanted to set the port speed you would need a managed switch.

However, today most unmanaged switches work just fine.

Can you setup a test where you copy from one VM to another VM within your VM server and see what the performance is?

Do you have multiple physical computers at that site?  If so, can you copy a file between the two and see what the performance is?


Author Comment

ID: 33697407
Ok I am convinced it is a port speed issue... But so what can I do given that we have an unmanaged switch?
Would it be enough just to set the port speed on the servers themselves (in Windows networking)?
LVL 57

Accepted Solution

giltjr earned 250 total points
ID: 33697599
What has convinced you that it is a port speed issue?

You can try setting the speed and duplex.

Typically with switches speed is negotiated fine, it is duplex negotiation that causes problems.

Duplex mismatch would show as decent speeds to start and then quickly drops off to next to nothing.
LVL 32

Assisted Solution

aleghart earned 250 total points
ID: 33698881
What kind of local switch do you have?  Even a small dumb gigabit switch should have sufficient fabric to handle a few servers and a 100Mbps WAN connection.  If you're running on a Fast Ethernet switch, could you be having problems with insufficient bandwidth on the switch?

If it's problems with remote clients...nothing you can control, test, replicate, etc. perhaps you can find somoene with a known fast connection that can give your server a run.  For instance, someone closer to you with a T1 or better.  With distant clients you are fighting against latency, lost packets, bad/slow routing, slow connectivity at the client's end.

Switching providers in your datacenter could also affect connectivity speed for foreign clients.  Their route to you may increase by several hops, and have to route through more congested networks before they get to your ISP.

Author Closing Comment

ID: 33835907
Thanks guys, things seem better now tho we are still doing tests.

Featured Post

Microsoft Certification Exam 74-409

Veeam® is happy to provide the Microsoft community with a study guide prepared by MVP and MCT, Orin Thomas. This guide will take you through each of the exam objectives, helping you to prepare for and pass the examination.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Creating an OSPF network that automatically (dynamically) reroutes network traffic over other connections to prevent network downtime.
Load balancing is the method of dividing the total amount of work performed by one computer between two or more computers. Its aim is to get more work done in the same amount of time, ensuring that all the users get served faster.
Here's a very brief overview of the methods PRTG Network Monitor ( offers for monitoring bandwidth, to help you decide which methods you´d like to investigate in more detail.  The methods are covered in more detail in o…
This video gives you a great overview about bandwidth monitoring with SNMP and WMI with our network monitoring solution PRTG Network Monitor ( If you're looking for how to monitor bandwidth using netflow or packet s…

777 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question