New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Logically, if it is port issue, you would experience same issue in Linux. Have you tried running Linux to test?
I also have Clouvider 10gbps port, running Linux, everything works as expected.
Have you tried download drivers directly from their website? If you are using Intel NIC, check this: https://www.intel.com/content/www/us/en/products/details/ethernet/500-controllers/82599-10-controllers/downloads.html
I m not sure it is same datacenter or what but i m using clouvider commands for testing speed from their provided site link iperf3 -c ash.speedtest.clouvider.net -p 5200
well i tried with multi threads and hope its correct , but sadly same result
https://prnt.sc/1wvkomt
Well , speed is perfect in linux according to their test but issue is in windows OS.
ok thanks , let me check that link. thank you
Obviously the magic here is that your version of windows is not bundled with cracked malware and acquired from the shadiest site possible for warez and using half the bandwidth for its own wants
Man i installed Trial version directly from Microsoft site and i mentioned this in one of my previous reply , please read again
Feels like you can try this: Purchase the license from Clouvider for one month and ask them to install the OS for you. Once the 10G speed is confirmed, replace the license with your own one and I don't believe that replacing the license will cause a speed drop.
Maybe pay for licensed windows with the provider for a month and see if that helps... if same speeds show then the host will help. Otherwise you know there is some issue with your install.
I feel that's the only way you will get to the truth.
Check your UDP with -b 10G added to iperf3 command line.
Otherwise, you should be testing with a minimum of 3 parallel streams (add -P3 to your iperf3 command) when testing TCP.
You likely need to increase the window size in iperf3 on windows.
You likely need to search for the windows option that reserves bandwidth that prevents you from hitting maximum.
Lastly, check out TCP Optimizer from speedguide.net for making optimization.
Unfortunately, you used the wrong flag there, -t10 means run for 10 seconds. Please check @TimboJones reply above for the correct flag and usage.
Cheers!
https://www.speedguide.net/downloads.php
Run TCPOptimizer to max
Well , i installed drivers from Intel site and now i m getting proper speed using iperf3, thanks @nanankcornering ,
but strangely speed is same as before 2~3Gbps when i test on browsers , i also checked speed with downloading a 10GB file from same datacenter ashburn from clouvider link but download speed is same (2~3Gbps)
https://imgur.com/IFCJmZd
https://imgur.com/O46Zgpu
is it correct now? @TimboJones @TheLinuxBug
`C:\Users\Administrator\iperf\iperf-3.1.3-win64>iperf3 -c ash.speedtest.clouvider.net -P3 -p 5200
Connecting to host ash.speedtest.clouvider.net, port 5200
[ 4] local port 50278 connected to port 5200
[ 6] local port 50279 connected to port 5200
[ 8] local port 50280 connected to port 5200
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 287 MBytes 2.41 Gbits/sec
[ 6] 0.00-1.00 sec 557 MBytes 4.67 Gbits/sec
[ 8] 0.00-1.00 sec 288 MBytes 2.41 Gbits/sec
[SUM] 0.00-1.00 sec 1.10 GBytes 9.49 Gbits/sec
[ 4] 1.00-2.00 sec 286 MBytes 2.40 Gbits/sec
[ 6] 1.00-2.00 sec 559 MBytes 4.69 Gbits/sec
[ 8] 1.00-2.00 sec 285 MBytes 2.39 Gbits/sec
[SUM] 1.00-2.00 sec 1.10 GBytes 9.49 Gbits/sec
[ 4] 2.00-3.00 sec 289 MBytes 2.43 Gbits/sec
[ 6] 2.00-3.00 sec 553 MBytes 4.64 Gbits/sec
[ 8] 2.00-3.00 sec 289 MBytes 2.42 Gbits/sec
[SUM] 2.00-3.00 sec 1.10 GBytes 9.49 Gbits/sec
[ 4] 3.00-4.00 sec 291 MBytes 2.44 Gbits/sec
[ 6] 3.00-4.00 sec 549 MBytes 4.61 Gbits/sec
[ 8] 3.00-4.00 sec 291 MBytes 2.44 Gbits/sec
[SUM] 3.00-4.00 sec 1.10 GBytes 9.49 Gbits/sec
[ 4] 4.00-5.00 sec 284 MBytes 2.39 Gbits/sec
[ 6] 4.00-5.00 sec 562 MBytes 4.71 Gbits/sec
[ 8] 4.00-5.00 sec 284 MBytes 2.39 Gbits/sec
[SUM] 4.00-5.00 sec 1.10 GBytes 9.48 Gbits/sec
[ 4] 5.00-6.00 sec 284 MBytes 2.38 Gbits/sec
[ 6] 5.00-6.00 sec 562 MBytes 4.72 Gbits/sec
[ 8] 5.00-6.00 sec 284 MBytes 2.38 Gbits/sec
[SUM] 5.00-6.00 sec 1.10 GBytes 9.48 Gbits/sec
[ 4] 6.00-7.00 sec 284 MBytes 2.38 Gbits/sec
[ 6] 6.00-7.00 sec 562 MBytes 4.71 Gbits/sec
[ 8] 6.00-7.00 sec 284 MBytes 2.38 Gbits/sec
[SUM] 6.00-7.00 sec 1.10 GBytes 9.48 Gbits/sec
[ 4] 7.00-8.00 sec 284 MBytes 2.38 Gbits/sec
[ 6] 7.00-8.00 sec 562 MBytes 4.72 Gbits/sec
[ 8] 7.00-8.00 sec 284 MBytes 2.38 Gbits/sec
[SUM] 7.00-8.00 sec 1.10 GBytes 9.48 Gbits/sec
[ 4] 8.00-9.00 sec 284 MBytes 2.38 Gbits/sec
[ 6] 8.00-9.00 sec 563 MBytes 4.72 Gbits/sec
[ 8] 8.00-9.00 sec 284 MBytes 2.38 Gbits/sec
[SUM] 8.00-9.00 sec 1.10 GBytes 9.48 Gbits/sec
[ 4] 9.00-10.00 sec 285 MBytes 2.39 Gbits/sec
[ 6] 9.00-10.00 sec 561 MBytes 4.70 Gbits/sec
[ 8] 9.00-10.00 sec 284 MBytes 2.39 Gbits/sec
[SUM] 9.00-10.00 sec 1.10 GBytes 9.48 Gbits/sec
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 2.79 GBytes 2.40 Gbits/sec sender
[ 4] 0.00-10.00 sec 2.79 GBytes 2.40 Gbits/sec receiver
[ 6] 0.00-10.00 sec 5.46 GBytes 4.69 Gbits/sec sender
[ 6] 0.00-10.00 sec 5.46 GBytes 4.69 Gbits/sec receiver
[ 8] 0.00-10.00 sec 2.79 GBytes 2.40 Gbits/sec sender
[ 8] 0.00-10.00 sec 2.79 GBytes 2.40 Gbits/sec receiver
[SUM] 0.00-10.00 sec 11.0 GBytes 9.48 Gbits/sec sender
[SUM] 0.00-10.00 sec 11.0 GBytes 9.48 Gbits/sec receiver
iperf Done.
C:\Users\Administrator\iperf\iperf-3.1.3-win64>`
weird , iperf3 test is fine now but actual speed is same as before
need more help to fix this issue.
Few days ago i tried but no luck , maybe i m not familiar with it.
It seems like you can only do 2-3Gbps per thread. The port is 10Gbps, obviously.
Do you really need 10Gbps when you download something in your browser? I have no idea how optimized the downloads are in major browsers, but I find it hard to believe that they are optimized for 10Gbps links.
Ofc i need 10Gbps speed and therefore i paid for it , otherwise i have one more 1Gbps server from other provider and i get full 1Gbps speed from that server , i was planning to cancel that 1gb server but sadly i m stuck with speed issue in clouvider server.
atleast 7~9Gbps should be ok but 2~3Gbps is useless which is 150~200MB/s.
and i m just testing speed in browser and with IDM (internet download manager) for confirm about speed.
For me, if iperf already hit 9G, then it already works. I shouldn't be worried.
Try doing speedtest.net, find 10G servers, see if it hits one. The issue might be the browser, or even the SSD/NVMe/HDD in your server.
Now, bring real production/user traffic to your server, and see if it handles it well.
Why is that? Any reason you need to download file at ~10gbps from browser?
Did you try another tools, such as wget or any download manager which support multiple threads
question was if you really need 10Gbps for SINGLE DOWNLOAD TASK IN YOUR BROWSER ON SERVER
because typical cloud server use cases are using more than one single logical connection/session to single receiver, and there are used more threads, more logical connections, more transmit/receive peers, and in such case your EFFECTIVE speed will be close to your iperf results (not dumb single thread browser test result)
if your use case is really only to download one single file with max 10G speed, then use more advanced multithreaded downloader tool (i.e. axel, aria2, aget)
From my experience, it's rare to find places that will serve you files at 10gbps (not google drive nor dropbox for example, ). Try OVH's test files, the largest one:
http://ovh.net/files/
If you get 400+ MB/s you should not worry much.
Edit: ofc don't download using chrome! use something smarter and more efficient.
You might try disabling TCP Checksum Offload (IPV4) & (IPV6) under device manager and advanced on the interface.
Please go back and read my first response to you in this issue. You likely again have the wrong expectations about how the internet works. MOST websites and file delivery services are NOT going to provide 10Gbit download over TCP / HTTP / HTTPS. For many reasons, the largest of them being:
You are assuming the drives in the source even have drives able to read data off them that fast -- unless you specifically know the server you are downloading from you can't even confirm that server can either provide disk read access that fast nor transport the file that fast to you.
Things that are not on your direct network will have overhead based on transit available on the pipes your traffic is going through -- Cloudvider is not promising you full 10Gbit speeds outside their network, they can't possibly do this as they don't manage those networks.
TCP and HTTPS especially have CPU overhead to them -- UDP used by iperf3 generally does not, so you may be able to fill the pipe with UDP packets fully but this will not reflect the speeds you will see over TCP where there is much more overhead on both sides of the transaction.
If you are able to pull > 1Gbit from a location then you are doing good as most services don't even provision many single servers with more than a gigabit port -- even VPS services here LET are mostly hosted on host nodes with 1Gbit connectivity only.
If you are going to get 10Gbit it is going to be in a multi-threaded environment from a service you know is running fast enough hardware on a well enough connected network to the one you are on -- based on your iperf3 you pasted last you are easily reaching 10Gbit in multi-threaded testing so you are getting what you are paying for -- you will just need to understand you are not going to get 10Gbit in a single transaction unless it is on the local Clouvider network and expecting that is not realistic. Anything over a gigabit from a server you don't know specifically can support 10Gbit should be considered good as most services won't even have more than 1-2 gbit connectivity in single server.
my 2 cents.
Cheers!
Load linux on the machine. If you getting the same results then you have evidence then. You gotta do some troubleshooting first and exhaust all efforts.
OP should throw a pizza party to thank everyone for all the free consulting.
All he knows that he has paid for 10Gbps, he should get 10Gbps from any source ... nothing else matters
maybe he is preparing Death Star with 10G cannon able to destroy every planet... I mean host
That is an absolutely stupid way to confirm download speed, since the limit is most likely in the browser!
Clouvider has proven that they deliver 10Gbps, the problem here is that you just don't know how to use it.
TL;DR OP have actually no idea what he is doing and everyone helping him deserves at least a treat.
If you have the RAM, also use a RAMdisk. I don't think you mentioned your disk setup, have you?