Network speed test throughput monitor

I am looking for a quick and dirty speed test to continously monitor the bidirectional throughput of a transponder and POE device while they are undergoing environmental testing.
Would be nice if it could log throughput, dropped packets, and bit error rates between two computers and did not crash when a fault causes one end to stop responding like the C program we have now.
Our C programmer looks at us like we are insane when we ask him to not stop display an errror if the client or server times out. 

Yes, I know how to handle and suppress errors in LabView. I was mainly looking for a good example on a network speed test.
The test we have uses was written by one of our software guys, it uses two computers in a client-server arrangement and stops and displays an error message when the client computer loses connection to the server. Regardless of the reasoning I have given the programmer refuses to change it because he wrote the test for him and that's how he wants it.
I want a program that can use one computer with two network cards. 
It will send a data stream out one NIC, through the UUT, and back in the other NIC.
Where it Will be checked for errors and dropped packets, and sent back through the UUT.
A continuous throughput speed measurment will also be logged along with errors, dropped packets, and retries.
I have never used any of the LV networking, tcp, etc  VI's and was hoping someone had done this before because I do not have time to start from scratch.

Similar Messages

  • Standard local network speed test?

    I recently setup up a dual-band (dedicated n, dedicated g) wireless network, and I think I am having some issues with throughput (qualitative observation). So I started trying to find a good metric for performance of the wireless systems. Note, I am talking here about the local wireless (Airport devices between home computers), not the connected internet speed (cable, DSL, fiber).
    I see a lot of people copying files and timing the transfer. Is this the best there is? From what I've read, ping isn't such a good metric for network throughput. You can watch the kb/s transfer report on copy, but that isn't exactly stable or indicative of general performance. Any other tools, programs, methods, etc.? Or should I just be timing file transfers?
    Thanks!

    Yes, I know how to handle and suppress errors in LabView. I was mainly looking for a good example on a network speed test.
    The test we have uses was written by one of our software guys, it uses two computers in a client-server arrangement and stops and displays an error message when the client computer loses connection to the server. Regardless of the reasoning I have given the programmer refuses to change it because he wrote the test for him and that's how he wants it.
    I want a program that can use one computer with two network cards. 
    It will send a data stream out one NIC, through the UUT, and back in the other NIC.
    Where it Will be checked for errors and dropped packets, and sent back through the UUT.
    A continuous throughput speed measurment will also be logged along with errors, dropped packets, and retries.
    I have never used any of the LV networking, tcp, etc  VI's and was hoping someone had done this before because I do not have time to start from scratch.

  • Network speed test in jsp

    I would like to know which APIs are necessary to develop a network speed tester in jsp or which method to do this, knowing about which APIs it's necessary, I will study them. Here is a example of a network speed tester that I'm trying to do
    Examples:
    http://www.coiinc.com/speedtest/initialmeter.php
    http://www.pcativo.com.br/testes.asp?teste_tipo=velocidade
    http://www.navegante.com.br/medidor/initialmeter.php
    http://medidor.interair.com.br/
    http://www.rjnet.com.br/velocimetro/index2.htm

    Sir,
    For at least a couple of these options it appears that the test is actually based on JavaScript.
    What seems to happen is that at the beginning of the page there is a JavaScript variable that is set with the time.
    Then there is a big amount of some sort of data. Which I would guess has been measured to be X big. Let's say 1 MB.
    Finally there is another bit of Javascript which again takes the time and then finds out the difference and from that one can sort of conclude the time difference.
    But I am not convinced this will always work perfectly but who knows... anyway your use of JSP wouldn't preclude you from just basically copying what these others are doing.
    Sincerely,
    Slappy

  • Network speed test under linux?

    I've been experimenting with jumbo frames on my gigalan network.  Basically, I set the mtu size on the two test machine, then transfer some files to and from them measuring the throughput.  Is there a program to test throughput between two machines on a network that doesn't involve the user literally copying files?

    I have used something in the past when I was also looking at jumbo frames until I realised my NIC doesn't support them. I think it was called lperf. IT was quite interesting to see the difference between TCP and UDP. There are a bunch of utilities for checking network throughput listed here.

  • Network speed monitoring?

    Hey,
    Does anyone know of any software that will monitor AND log my network connection information (throughput in/out, specifically)? I've found plenty of programs that will monitor the connection, but will not save the data over time.
    I want to do this because I'm suspicious of my ISP... I have the " up to 18mb/s" service, and various internet speed tests ROUGHLY corroborate this speed, however it seems like almost every website I go to (except for the speed test websites LOL) loads incredibly slow.
    Ultimately I'd like to have a printable graph of my network throughput over a specified period of time (days, weeks, months, even years).
    Anyone have any ideas?

    Hi Greenwall, and a warm welcome to the forums!
    Does anyone know of any software that will monitor AND log my network connection information (throughput in/out, specifically)?
    http://homepage.mac.com/rominar/net.html
    I want to do this because I'm suspicious of my ISP... I have the " up to 18mb/s" service, and various internet speed tests ROUGHLY corroborate this speed, however it seems like almost every website I go to (except for the speed test websites LOL) loads incredibly slow.
    See if this speeds it up...
    Try putting these numbers in Network>TCP/IP>DNS Servers, for the Interface you connect with...
    208.67.222.222
    208.67.220.220
    Then Apply

  • How to test network speed

    How do I test the speed of my network to determine if Time Capsule or my Internet provider is behind a real slowdown in my internet access. I ran the java test at DSLReports.com and got the following results:
    Compared to the average of 971 tests from comcast.net: download is 98% worse, upload is 98% worse.
    I'm unsure where to look or what to look for in trying to find a cause for the problem but wanted to at least try and make a stab at finding out before I called Comcast.

    You could try connecting a computer directly to your broadband modem and run the test again.  That would remove the Time Capsule from the configuration.  You may need to reset the modem to convince it to connect to a "new" device.
    Are you connecting to the Time Capsule via Ethernet or Wi-Fi?  If by Wi-Fi, the quality of your Wi-Fi connection could have a big affect on the network speed.  In that case, use the advice in the second message of this thread to investigate the signal and noise levels that your Mac sees:

  • Network speed affected by large file copy operations. Also, why intermittent network outages?

    Hi
    I have a couple of issues on our company network.
    The first is thate a single large file copy imapcts the entire network and dramatically reduces network speed and the second is that there are periodic outages where file open/close/save operations may appear to hang, and also where programs that rely on
    network connectivity e.g. email, appear to hang. It is as though the PC loses it's connection to the network, but the status of the network icon does not change. For the second issue if we wait the program will respond but the wait period can be up to 1min.
    The downside of this is that this affects Access databases on our server so that when an 'outage' occurs the Access client cannot recover and hangs permamnently.
    We have a Windows Active Directory domain that comprises Windows 2003 R2 (soon to be decommissioned), Windows Server 2008 Standard and Windows Server 2012 R2 Standard domain controllers. There are two member servers: A file server running Windows 2008 Storage
    Server and a remote access server (which also runs WSUS) running Windows Server 2012 Standard. The clients comprise about 35 Win7 PC's and 1 Vista PC.
    When I copy or move a large file from the 2008 Storage Server to my Win7 client other staff experience massive slowdowns when accessing the network. Recently I was moving several files from the Storage Server to my local drive. The files comprised pairs
    (e.g. folo76t5.pmm and folo76t5.pmi), one of which is less than 1MB and the other varies between 1.5 - 1.9GB. I was moving two files at a time so the total file size for each operation was just under 2GB.
    While the file move operation was taking place a colleague was trying to open a 36k Excel file. After waiting 3mins he asked me for help. I did some tests and noticed that when I was not copying large files he could open the Excel file immediately. When
    I started copying more data from the Storage Server to my local drive it took several minutes before his PC could open the Excel file.
    I also noticed on my Win7 client that our email client (Pegasus Mail), which was the only application I had open at the time would hang when the move operation was started and it would take at least a minute for it to start responding.
    Ordinarlily we work with many files
    Anyone have any suggestions, please? This is something that is affecting all clients. I can't carry out file maintenance on large files during normal work hours if network speed is going to be so badly impacted.
    I'm still working on the intermittent network outages (the second issue), but if anyone has any suggestions about what may be causing this I would be grateful if you could share them.
    Thanks

    What have you checked for resource usage during one of these copies of a large file?
    At a minimum I would check Task Manager>Resource Monitor.  In particular check the disk and network usage.  Also, look at RAM and CPU while the copy is taking place.
    What RAID level is there on the file server?
    There are many possible areas that could be causing your problem(s).  And it could be more than one thing.  Start by checking these things.  And go from there.
    Hi, JohnB352
    Thanks for the suggestions. I have monitored the server and can see that the memory is nearly maxed out with a lot of hard faults (varies between several hundred to several thousand), recorded during normal usage. The Disk and CPU seem normal.
    I'm going to replace the RAM and double it up to 12GB.
    Thanks! This may help with some other issues we are having. I'll post back after it has been done.
    [Edit]
    Forgot to mention: there are 6 drives in the server. 2 for the OS (Mirrored RAID 1) and 4 for the data (Striped RAID 5).

  • I am desperate!! slow choppy network speeds (not actual inability to cnnct)

    I have a 13" macbook air that has the following problem: I can connect to wifi networks (including at home WPA and also tried WPA2 settings) but even though I may get proper speed at first, I soon lose that and it becomes much slower and choppy (if I open activity monitor I can see the speed going up and down like crazy...). My internet speed at home is 12mpbs set by my provider. When I perform speed tests (vis speedtest.net or downloading files from other speed test sites) on my windows PC that is also on this network I get the full download speed. However with this MAC while I initially did get the full speed, it is now down to an average of 3-5mbps and the activity monitor, again, shows choppy behavior so it will be at 8 then at zero almost and then up and down again but never near the full speed. I tried to connect to other networks and I have the same problem - once this happens it stays. I tried removing the airport connection in network preferences and adding it again it does not help.
    I TALKED TO APPLE AND HAD MY COMPUTER TESTED BY THEM - they didn't find any hardware issues (they said it was fine - but it isn't) but they still replaced the airport hardware in the computer - both the chip and the antenna (which meant replacing the entire screen). I brought the computer home, ran a speed test and everything was fine. I thought GREAT but then something happened and the problem returned. What I did (which I think was unrelated) was connect via SMB to a PC at home and I tried to stream a movie file. I noticed it kept 'pausing' so I opened up activity monitor again and to my HORROR I saw the choppy wave with network speed very inconsistent going up and down like crazy... I tried connecting to my slingox using the sling web access after disconnecting from my home PC but got partial speed and lots of drops too. I tried the speed test again multiple times, nothing seems to work.
    SO I don't know what to do at this point. I have 10.6.6 running. I heard something about a firmware update but I am new to mac so not sure how to even do that or what else I can do to solve this problem or what is causing it. BTW - logging in now from the guest account still does not fix the issues and they remain the same.
    I am desperately looking for help!!
    Thanks a lot!

    Have you been able to reproduce your problems on other wireless networks other than at your house? It appears you are saying that this occurs on all wireless, but I'm not sure from your comment. If it's only at home, it would appear to be a compatibility issue between your wireless access point and the Macbook Air, which isn't uncommon. Sometimes a firmware upgrade to the access point can fix it, sometimes it might require switching to a different brand of access point. It would be great if you could borrow (or buy and return) a different access point/router to test. An Apple airport would be the best to use of course. Not the easiest solution, but lets you help rule out a hardware issue with your air.
    My isp provides 15mbps but my wireless devices such as my iphone, ipad, laptop usually only get 6.5-8 mbps when testing speeds to the internet. I don't expect it to be as good as the wired due to interference, signal, etc. I don't get the choppy behavior however. Firware updates from Apple will appear in the same manner as standard software updates. If there is one available for your system it should appear when you check for software updates. That being said, I believe you can also manually download them from the Apple site.

  • Upgrade to Snow Leopard 10.6 has downgraded networking speed

    I have 3 computers on my 100% wired Gigabit network at home. All 3 computers are connected to the same router.
    1 is a linux machine, 1 is a Macbook (C2D 2.4GHz 4MB RAM) running Leopard 10.5.8, and 1 is my iMac (C2D 2.4GHz 4MB RAM) running Snow Leopard 10.6.1. Both iMac and Macbook have identical specs and RAM, only difference is one is running 10.5.8 and one is running 10.6.1.
    Prior to upgrading my iMac to SL, I'd consistently get between 10-20Mbps download result using dslreport.com speedtests. I'm on Comcast. Now after upgrading to SL on the iMac, I can only achieve around 2.5Mbps maximum speed test results. I performed this test 5x over several days to make sure there wasn't some coincidental anomaly. As a test I performed the same speed test on the Macbook and it indeed hit above 15Mbps speed result consistently 5x in a row and over several days. My linux machine also achieves the same 15+Mbps speed result.
    I've tried tweaking the iMac's SL network settings, but nothing helps. What has happened in SL to degrade network throughput? I can clearly and consistently reproduce this issue.

    Aha! I've found the problem. Its the Flash 8 speed test that is the problem on SL. When performing the same speed test using the Java speed test, I achieve the 15+Mbps download speeds.
    Flash 8 speedtest
    Sun Java plugin speedtest
    I've done the test 5 times in a row and I get the same results form each respective speedtest type.
    So the problem is what is wrong with Flash on Snow Leopard?

  • How does Activity Moniter measure Network speeds.

    How does Activity Monitor measure network transfer rates? They are always much lower than Web based speed checks. Typically my download speed on Activity Monitor is 162 KBS whereas Web based speed check return 1.2 MBS.

    Duh! I should have realized the AT&T and Activity Monitor are using two different measurements. AT&T uses bits. Activity Monitor uses Bytes. Multiply Bytes by 8 to get bits. So 162 K-bytes in Activity Monitor = 1.29 M-bits in AT&T speed test. The problem comes in misreading the labeling of the units measured. Which is something that causes confusion frequently because the industry is not uniform in its use of labels.
    It is interesting that AT&T uses the higher number. Is this a marketing strategy? Don't get me wrong, I am very satisfied with their service.
    I am also VERY satisfied with my MacBook Pro. The best Mac I ever owned (since my first Mac in 1993 and Apple 2's for 10 years before that). Awesome system software, boots faster than previous MacBooks I owned. Has been flawless in performance. I bought it from Apple refurbished. It came with a larger HD than I ordered. I have used both WinTel (since 1993) and Apple machines. There is no comparison IMHO.

  • [SOLVED] Network speed very slow, no apparent reason

    Hello, recently I switched from Windows 7 RC1 to Arch on my home machine (I have used Gentoo, Ubuntu, and Fedora 10 on the same machine in the past), and for some reason network speed is very slow.
    I have Verizon's 20mbit/5mbit package, and I have always gotten that speed.
    Using speedtest.net and 100mb.test from cachefly on multiple computers I have come to the conclusion that it is infact my Arch install that is causing the problem:
    - All other machines on my network are getting 20/5 (both wired and wireless)
    - I ran a speedtest from 2 other machines using the ethernet cable that this PC is on.  Again, full 20/5.
    - scp transfer of 100mb.test from this PC -> other Arch box: ~2.8MB/s.
    - scp transfer of 100mb.test from other Arch box -> this PC: ~2.8MB/s.
    - scp transfer of 100mb.test from this PC -> UK VPS (100mbit line): ~539KB/s
    - scp transfer of 100mb.test from UK VPS -> this PC: ~76KB/s
    - scp transfer of 100mb.test from this PC -> Chicago Server (dual gbit lines): ~563KB/s
    - scp transfer of 100mb.test from Chicago Server -> this PC: ~91KB/s
    speedtest.net result:
    Upload speed seems to be unaffected.
    I have tried disabling TCP window scaling, and appending my hostname to /etc/hosts.
    What is weird though is that the other Arch box has an identical network config,  I don't see any reason why it shouldn't work.
    My NIC is an integrated Realtek something, I can get the exact model if needed.
    Last edited by whipsch (2009-06-05 18:29:21)

    Hi, whipsch
    Can you try ethtool to see if your ethernet card is actually negociating and using a 100Mb full duplex link on your LAN ? If not you can try to force the link parameter with ethtool. Also maybe the driver of your NIc has options related to link negociation.
    Hope this helps,
    JF

  • The Delete Speed Test for Pages 2

    Hello Pages fans,
    “Speed” is a subjective term, so I wonder what people will find if they do this test:
    Paste a phrase such as “The quick brown fox jumped over the lazy dog’s back. ” ten times to make a dummy paragraph. Put the cursor at the end of the paragraph and press and hold the delete key and count the seconds it takes to delete back through the the entire paragraph, letter by letter. (This will depend on the key repeat rate you set in System Preferences.)
    If I delete a particular paragraph this way using Pages 2, when viewing the document in narrow format (i.e. viewing “fit page”) it takes about 22 seconds. If I view “fit width” and stretch the width across my 23-inch cinema display the same task takes about 101 seconds.
    Using Pages 1 it takes about 22 seconds (“fit page”) or about 27 seconds (“fit width”). Not as bad.
    I have tried every single fix proposed in this discussion forum to speed up Pages - to no avail. I am using a Mac Mini 1.42 GHz with 1GB RAM, a 23-inch display, 10.4.5 and the very latest updates. I have no other problems whatsoever in any other applications.
    Incidentally, In Word 2004 the same exact test takes about 17 seconds no matter what (whether viewed at 10%, 500% or anything in between).
    Trust me- I don’t actually sit around deleting paragraphs like this all day, but I do notice a very annoying typing lag. I’ve basically given up on Pages as a serious writing tool because of this typing lag / speed issue and gone back to Word. Pages 1 works fine and served my purposes for a year, but what future does the Pages application have if version 2 performs so poorly?
    Do those who report having blazing fast Pages 2 performance experience this problem with this test?
    Yes, this is a dreadfully boring test! But at least it’s not subjective and it’s not as iffy as watching the system monitor percentages jumping around.
    Thanks in advance for any responses.
    various macs...   Mac OS X (10.4.4)  

    True, MMINSO, I can just go use some other word processor. (Thanks for the tip.) I'm sorry to laugh at your suggestion but you entirely missed the point of my keyboard input test when you told me to just delete the whole paragraph at once.
    Ee gads... we could start a flame war over this! Can anyone think of a better, more reliable speed test to measure how Pages handles keyboard input?
    When I type text while viewing a document at full width across my screen, the typing is much slower than when I view the document tiny size. The typing lags. By "lags" I mean that the keyboard input is behind what I am typing. I devised a mind-numbingly simple little keyboard input test to replicate this problem perfectly. Again: This test directly replicates the type of keyboard input slowdown problem I encounter while doing normal typing. So it's certainly relevant, and it's also measurable and replicable.
    Naturally, it's not the actual number of seconds that counts, it's whether any given keyboard input task slows down when we do it viewing at different widths. The keyboard input rate changes, you see.
    I had hoped that someone would verify whether they experience the same lag in typing under the same conditions. In other postings, others have simply described the problem as "infuriating" and dropped Pages entirely. I don't think that's very useful. I was trying to be a bit more scientific about it in the hopes that it might actually be helpful.
    Maybe there's a better way to accurately demonstrate keyboard input lags in Pages?
    Maybe my machine is the only one experiencing any slowness.

  • How to track network speeds - maybe through SQL or open to others

    The network speed is dramatically slower to the SAN, but it also makes me wonder how slow things are between the Webserver and the external world.
    What's the best way to monitor what-I-assume are network delays and which part of the networks are causing the problem.
    -- Needs to be run on the VM Machine xxxx to work
    /* Local backup */
    BACKUP DATABASE [play_pen] TO
    DISK = N'x:xxxx\testbackup.bak' WITH COPY_ONLY, FORMAT, INIT,
    NAME = N'Speed backup', SKIP, NOREWIND, NOUNLOAD, STATS = 10, CHECKSUM, CONTINUE_AFTER_ERROR
    -- BACKUP DATABASE successfully processed 35289 pages in 7.638 seconds (36.095 MB/sec).
    /* Intra-Network backup*/
    BACKUP DATABASE [play_pen] TO
    DISK = N'\\xxxx\testbackup.bak' WITH COPY_ONLY, FORMAT, INIT,
    NAME = N'Speed backup', SKIP, NOREWIND, NOUNLOAD, STATS = 10, CHECKSUM, CONTINUE_AFTER_ERROR
    -- BACKUP DATABASE successfully processed 35289 pages in 14.152 seconds (19.480 MB/sec).
    /* SAN backup*/
    BACKUP DATABASE [play_pen] TO
    DISK = N'\\xxxx\testbackup.bak' WITH COPY_ONLY, FORMAT, INIT,
    NAME = N'Speed backup', SKIP, NOREWIND, NOUNLOAD, STATS = 10, CHECKSUM, CONTINUE_AFTER_ERROR
    -- BACKUP DATABASE successfully processed 35289 pages in 55.106 seconds (5.002 MB/sec)

    Hi,
    You can use the tools like Resource Monitor and Network Monitor.
    Resource Monitor is a tool that you can use to monitor the usage of CPU, hard disk, network, and memory in real time. You can open it by clicking the
    Start button. In the search box, type Resource Monitor, and then, in the list of results, click
    Resource Monitor.  If you're prompted for an administrator password or confirmation, type the password or provide confirmation.‌
    Meanwhile, I think it is proper to dig the networking issue on the Windows Server forum via the below link.
    http://social.technet.microsoft.com/Forums/en-US/home?forum=winservergen
    Additional information:
    Using Network trace to troubleshoot intermittent connectivity issues
    http://blogs.msdn.com/b/dataaccesstechnologies/archive/2010/06/18/using-network-trace-to-troubleshoot-intermittent-connectivity-issues.aspx
    Thanks.
    Tracy Cai
    TechNet Community Support

  • Slow Speed Tests

    Hay everyone I have an AirPort extreeme located in my front room about 15 feet from my bedroom. In the bedroom I have an iMac 27" mid-2011 model. I have been running speed tests from speedtest.net and have been getting very slow download rates, any where betwee 1.4 and 5.0 mbps. I pay for a 25 mbps speed. I have full signal strength everywhere in the house the speed is just terrible. Does anyone know how I can increses the speed of that computer. I have changed the channel to one that is not used as much in my area, as I looked at iStumbler. Also if I sit in the room next to the router I get 10 - 15 mbps. Any help would be amazing. Thanks.
    PS I also have an AirPort express set to extend the network in the bedroom where I am having the slow speed.

    I'm not able to provide a solution though I too find that the AE is not as fast as it is supposed to be. At one time I had a Belkin Wireless and an ADSL connection that topped out at 1.5MB/sec. Now I have a 50MB/sec internet connection and honestly, I have seen an increase in my speed over wireless but the Belkin really seemed quite zippy! I play World of Warcraft and I remember someone saying to me that AE or our Macs have Firewalls set-up that we don't know about or "something" that limits the speeds of the connections - which PCs and non-Apple products do not. Constantly see when downloading new patches from WOW, messages that I am behind a firewall - when I do not have firewall set on the AE nor on my Mac...

  • OVM 3.0 - network speed

    Hi together,
    quick question regarding domU network throughput in OVM 3.0.3.
    We did a very simple TCP Network Throughput Benchmark with NETIO and ended up with the following result.
    Setup:
    - OVM Server 3.0.3 on x86_64
    - OVM Guests OEL 5.7 64bit (2.6.32-200.13.1.el5uek) PVM with xennet
    - 10Gbit physical network (especially for the bonds serving the VM network bridges)
    Network Speed between two different dom0`s:
    Packet size 1k bytes: 320.12 MByte/s Tx, 302.97 MByte/s Rx.
    Packet size 2k bytes: 367.94 MByte/s Tx, 472.63 MByte/s Rx.
    Packet size 4k bytes: 694.60 MByte/s Tx, 841.67 MByte/s Rx.
    Packet size 8k bytes: 953.76 MByte/s Tx, 972.38 MByte/s Rx.
    Packet size 16k bytes: 1012.60 MByte/s Tx, 1.017 GByte/s Rx.
    Packet size 32k bytes: 1.002 GByte/s Tx, 1.008 GByte/s Rx.
    Network Speed between two domU´s  running on the same dom0:
    Packet size 1k bytes: 354.65 MByte/s Tx, 382.91 MByte/s Rx.
    Packet size 2k bytes: 468.07 MByte/s Tx, 473.96 MByte/s Rx.
    Packet size 4k bytes: 495.54 MByte/s Tx, 575.83 MByte/s Rx.
    Packet size 8k bytes: 526.07 MByte/s Tx, 548.81 MByte/s Rx.
    Packet size 16k bytes: 533.61 MByte/s Tx, 603.46 MByte/s Rx.
    Packet size 32k bytes: 571.73 MByte/s Tx, 650.30 MByte/s Rx.
    Network Speed between two domU´s running on different dom0`s:
    Packet size 1k bytes: 155.06 MByte/s Tx, 156.07 MByte/s Rx.
    Packet size 2k bytes: 176.94 MByte/s Tx, 196.74 MByte/s Rx.
    Packet size 4k bytes: 185.37 MByte/s Tx, 200.06 MByte/s Rx.
    Packet size 8k bytes: 157.51 MByte/s Tx, 218.15 MByte/s Rx.
    Packet size 16k bytes: 191.40 MByte/s Tx, 202.36 MByte/s Rx.
    Packet size 32k bytes: 185.55 MByte/s Tx, 204.82 MByte/s Rx.
    As you can see, the domU`s on two different OVM Servers have arround 80% performance loss compared to the available speed on the dom0 layer.
    2 domU´s on the same dom0 are around 50% slower.
    Is that normal?
    If not, what can be tweaked?
    Some reference measurements on other environments would be great.
    Thanks,
    Claudius

    I had a similar slowdown due to dropped packets (and the necessary retransmissions) using a private network on a single host. See thread: Re: Private network dropping packets
    It resulted in a bug but there was a workaround. I'd do an 'ethtool -K eth0 rx off tx off sg off tso off' and then an 'ethtool -K eth0 rx on tx on sg on tso on'.
    a) the second command undoes what the first command does,
    b) my network path does not include eth0 in the link that is dropping packets!
    c) the commands must be repeated after any host reboot.
    I have no idea what might be the underlying cause - my guess is an obscure race condition in the Oracle xen network stack - so I have no idea if this workaround will affect your issue. You can try the pair of commands on both hosts (pick any interface) and see what happens. You also might look at the traffic to see if dropped packets are the problem and pin down where they are being dropped - that will be tedious!
    Good Luck!

Maybe you are looking for