Resource Monitor TCP Connections High Latency

Hi,

Wonder if anyone can help with a problem I'm seeing on one of our SAP servers running Windows 2012 and SQL 2012, we are seeing high Latency in Resource Monitor under Network - TCP Connections.  The Connections are ranging from 200 - 300ms but when performing network tests we are getting <1ms response.

Is the TCP Connections taking into account how long that service takes to respond IE SQLSERVR.EXE is taking 288ms to respond it could be related to that processes being very busy so it is a CPU bottleneck or is it purely a network related issue?

Thanks
Jamie786Asked:
Who is Participating?

[Product update] Infrastructure Analysis Tool is now available with Business Accounts.Learn More

x
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Zephyr ICTCloud ArchitectCommented:
Do you see CPU spikes at the moment the service responds to queries? That would answer your question quickly, if it is network related it could be different things, e.g. mtu, fragmentation, duplex-mismatch, congestion, ...

I'm guessing your server has 1Gb ports, how is it connected? Switch? What's the info on the switch? 1Gb ports? what are the ports' statistics?
0
Jamie786Author Commented:
These are connected via a 10Gb port and go directly into a switch, I don't see CPU spikes that match the high latency as the latency is always high IE always at 288ms for instance and the CPU is rather low IE about 10% usage.

a simple ping from the server to a desktop and back is showing as <1ms all the time shall we say.  If it was a duplex-mismatch or congestion I would expect to see a standard constant ping taking a lot longer to respond.
0
Zephyr ICTCloud ArchitectCommented:
yeah probably but I can't make assumptions,, so I always ask all the questions I can think of at the time I start answering :)

Interesting, 10Gb ports but still see high latency, might it be that the congestion is elsewhere in the network path? A switch in between that can't handle the traffic well? What happens when you ping with large packets, do you see the same latency or not? What about using the don't fragment bit?

E.g:
ping -l 1500 and ping -l 9000

Ping -l 1500 -f
0
10 Tips to Protect Your Business from Ransomware

Did you know that ransomware is the most widespread, destructive malware in the world today? It accounts for 39% of all security breaches, with ransomware gangsters projected to make $11.5B in profits from online extortion by 2019.

Jamie786Author Commented:
I can ping all the way upto 65426bytes and the most it goes upto is 2ms, when I set don't fragment switch it accepts and responds in <1ms...
0
Jamie786Author Commented:
Sorry When I use ping -l 1500 -f I get Packet needs to be fragmented but DF set
0
Zephyr ICTCloud ArchitectCommented:
yes, that's normal ... It means that your packets can't be larger than 1500 bytes, you can find the limit by lowering the bytes, for example ping -l 1450 -f ... Until you get a reply, it can end at 1464 or something ...

Anyway, that's not really important, but it does tell me the mtu is probably 1500 on the switches/routers ... Are you doing these tests on the server to a client, from a client to the server or??

How much traffic are we talking about? Large packets or small bursty packets/traffic?

Have you considered the latency can come from something else? For example disk access being slow? Has that been looked at? If SQL can't lookup the needed data fast enough on the disks it might also cause latency?
0
Jamie786Author Commented:
The disks do spike Latency wise but the average is below 20ms.... We are seeing various locks within sql is someone runs a large sql task but this latency that I can see if so constant IE when all other resources are more or less completely free I'm still seeing high latency within TCP Connections...

Do you know how TCP Connections works out the latency so I can try to replicate the same results elsewhere?
0
Zephyr ICTCloud ArchitectCommented:
well ... TCP reacts on latency differently according to the issues it undergoes ...

I would try to get to the bottom of it using PsPing from the sysinternal tools... It can give you more information on latency and bandwidth.

Also, try to work out what is your path, what nodes are you passing between your test point and the server, if any.
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Jamie786Author Commented:
good idea to use PsPing, from useful utility.
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Windows Server 2012

From novice to tech pro — start learning today.