20

ISPs often prioritize traffic to speedtest.net so that they can brag how fast their connections are, while in reality, they don't provide that much bandwidth. They're perfectly aware that most users will only check that site for confirmation. You also have to keep in mind that transfer speed relies both on the client and the server. In today's world most ...


14

In addition to the other reasons posted, TCP connections don't work well with large files when the bandwidth-delay product becomes large. Like on an otherwise fast connection to an island. See Wikipedia's entry on TCP tuning. So Speedtest can dump a small file through the connection at 95 mb/sec, but wget can only get 10 mb/sec on a 20 MB file.


7

wget give good practical measure of the speed. The tests of Speedtest probably include kind of parallelism which can explain higher numbers. For good average speed test I think the time for download should be at least 90-120 seconds (to get good average)


3

One reason could be that often the maximum speed cannot be reached by just a single TCP connection. Speedtest.net recently introduced a single connection mode. Try this and see if it makes a difference. Then, for the download use for example aria2 with parameters to use multiple conections and compare. e.g. aria2c -d /dev -o null --allow-overwrite=true --...


3

I've also been searching for clarification here and this is what I've found. There is no polling period for this counter. It is incremented as discards happen. It does reset on reboot. I have found no documentation stating this, but have seen no exceptions to this in practice. It is not reset over any time interval. When the counter reaches the max value ...


3

Rather than having a client with the main function, you can have a service for that. For example, import ballerina/http; import ballerina/io; endpoint http:Listener passthroughEP { port:8243 }; endpoint http:Client clientEP { url:"http://localhost:9218", proxy: { host:"localhost", port:9219 } }; @http:ServiceConfig {...


3

The problem is simply this. Shared hosting. cPanel runs on its own in a separate process, not tied to Apache. If they're not using stuff like CloudLinux to limit the amount of CPU/memory users can use, this is most likely the sign you're on an overcrowded server. You get this in shared hosting all the time. I've worked in that industry for more than 5 years ...


2

"Load Average" is not natively supported in Windows-based servers. Check this thread - https://social.technet.microsoft.com/Forums/lync/en-US/e5fd5657-726a-4cf0-8d6f-d760bf26ec77/windows-equivalent-to-to-linux-load-average-15-minutes?forum=winserverpowershell Use PowerShell Get-Counter or WMI - Processor Queue Length in System ...


2

On Ubuntu 16.04 I just ran into this same issue. After some grousing around I found I needed to enable the service by editing the file /etc/default/sysstat Change ENABLED="false" to true: # Should sadc collect system activity informations? Valid values # are "true" and "false". Please do not put other values, they # will be overwritten by debconf! ENABLED="...


2

Additionally, you can run proxy_server.bal which you are referring as 2 and configure the JMeter in order to send requests via Proxy Server as follows. jmeter -H my.proxy.server -P 8000 -u username -a password -N localhost Please refer this for more information on JMeter configuration.


2

It can. You can track IOWAIT and 'steal CPU' with htop (when you enable detailed stats), or track it with a tool like Munin. This is how you enable the detailed stats in htop: And this is a Munin example of a somewhat overloaded backup server, causing a lot of IOWAIT: How to set up Munin is described well on the net.


1

Use https://fast.com/ , this is a Netflix based speed test meaning it cannot be differentiated by ISP's from Netflix itself. This is a more accurate test than any other test generally. People won't be worried about how fast a webpage loads, but rather how quickly the videos buffer due to the increased bandwith necessary to display a video. ISP's often ...


1

You can find proxy configuration for JMeter on "Advanced" tab of the HTTP Request sampler where you can specify host, port and credentials If you have > 1 HTTP Request sampler you can set proxy for them all in one place using HTTP Request Defaults configuration element, this way you can set up default configuration for all HTTP Request samplers in HTTP ...


1

Turn your main program into an HTTP service first. In order to achieve the proxy behaviour, use forward client action to forward the inbound request to the backend service. Then you can perform the load test by invoking the the resource via JMeter request.


1

This doesn't actually answer the actual question, but I think this should help with solving the actual problem. The problem might well be in your web shop software, which simply is slow when there is enough traffic. You need to look at your traffic and check if high traffic ccorrelates with slowness. If there is correlateion, then it is the software that ...


1

You could use SNMP trap to get information of the server, it is completely free. It could be added in the add features wizard (SNMP Service). Once you add it go to services.msc find the SNMP service and configure it (if you want can change the user run it), under the tab security create the SNMP community a readonly one will really sufficient, googling for ...


Only top voted, non community-wiki answers of a minimum length are eligible