Measuring Shared Linux Hosting server performance

[tolong1]

I have Linux shared hosting. I signed up recently during Thanksgiving sales. Trying to set now the FTP speed is very slow.

Mine is not mission critical. I mainly host some static websites for my family business ( 3 domains) and few emails.

Total 15 email addresses for 3 domains. (support, sales, purchase, myemail my dad) so total 15 emails.



Request your help regarding checking server performance. What parameters should I check? How do I check?


Thank you

 

[davidktw]

I have Linux shared hosting. I signed up recently during Thanksgiving sales. Trying to set now the FTP speed is very slow.

Mine is not mission critical. I mainly host some static websites for my family business ( 3 domains) and few emails.

Total 15 email addresses for 3 domains. (support, sales, purchase, myemail my dad) so total 15 emails.



Request your help regarding checking server performance. What parameters should I check? How do I check?


Thank you

If yours a Virtual Private Server (VPS) Plan. There will be quite a number of performance measurement from using sysbench for cpu/mem/io, fio for disk(i/o), iperf/iperf3 for network. Unless you have something to compared to, these numbers tells little. However when it comes to web response, we use metrics like web requests per sec (RPS/***), web requests latency, time to first byte. This are closer to what end-user will experience when it comes to websites. But a lot of these performance can be outside of the server too, because if you are using external libraries sourced from CDN or other external sites, those matters as well. If your site is client rich and has alot of javascript/stylesheet or complicated client-side rendering, the end-user may perceive different user experiences. If you did not make use of caching well, that also contributes.

In your use case, you may get away with using hosting your static site in AWS S3 at a very affordable cost, much much lower than having a VPS. You can also engage the use of a free CDN like Cloudflare to provide some performance gain. Hence your end-users will be serving your static website as follows

BROWSER <--> CLOUDFLARE <--> AWS S3​

You can also choose GitHub if you like to host websites (https://docs.github.com/en/pages/co...es-site/about-custom-domains-and-github-pages)

Email wise, you can continue using your existing hosting plan if it provides emails too. Other alternatives will be getting your own domain, and attach it to Google Workspace or other such services.

 

[tolong1]

If yours a Virtual Private Server (VPS) Plan. There will be quite a number of performance measurement from using sysbench for cpu/mem/io, fio for disk(i/o), iperf/iperf3 for network. Unless you have something to compared to, these numbers tells little. However when it comes to web response, we use metrics like web requests per sec (RPS/***), web requests latency, time to first byte. This are closer to what end-user will experience when it comes to websites. But a lot of these performance can be outside of the server too, because if you are using external libraries sourced from CDN or other external sites, those matters as well. If your site is client rich and has alot of javascript/stylesheet or complicated client-side rendering, the end-user may perceive different user experiences. If you did not make use of caching well, that also contributes.

In your use case, you may get away with using hosting your static site in AWS S3 at a very affordable cost, much much lower than having a VPS. You can also engage the use of a free CDN like Cloudflare to provide some performance gain. Hence your end-users will be serving your static website as follows

BROWSER <--> CLOUDFLARE <--> AWS S3​

You can also choose GitHub if you like to host websites (https://docs.github.com/en/pages/co...es-site/about-custom-domains-and-github-pages)

Email wise, you can continue using your existing hosting plan if it provides emails too. Other alternatives will be getting your own domain, and attach it to Google Workspace or other such services.

Thank you @davidktw Appreciate.

Mine is not VPS. I am using shared web hosting from https://www.resellerclub.com/shared-hosting
At present I do not plan to go for database based web hosting but may consider in future using wordpress.

Basically we need some business email. so this hosting serves . I am also not a very technical person to install or configure cpanel or spam prevention etc. So this web hosting charges US$3 per month for the PRO plan I bought during the Thanksgiving sale. So Basic Static hosting and EMail meet the requirements.

Amazon AWS may give better performance but I need to manage all email hosting routing etc which may need some learning curve. Of course great support and mentorship of kind people like you will help me to speed up my learning curve.

The support team at Resellerclub.com asked me to verify the server performance using https://gtmetrix.com
Is this correct approach?

I am at present in the process of learning Amazon AWS for backup of myfiles around 70GB. Not decided yet Considering Amazon Glacier. I learned through painful way of losing some data. Stored some old data in Western Digital Hard Disk. Now unable to read. So I have to get out of all this Hard disk storage and move to Amazon Storage/Backup.

Thank you

 

[davidktw]

Thank you @davidktw Appreciate.

Mine is not VPS. I am using shared web hosting from https://www.resellerclub.com/shared-hosting
At present I do not plan to go for database based web hosting but may consider in future using wordpress.

Basically we need some business email. so this hosting serves . I am also not a very technical person to install or configure cpanel or spam prevention etc. So this web hosting charges US$3 per month for the PRO plan I bought during the Thanksgiving sale. So Basic Static hosting and EMail meet the requirements.

No problem, you use what you need, but you don't have to limit your options to just one.

Amazon AWS may give better performance but I need to manage all email hosting routing etc which may need some learning curve. Of course great support and mentorship of kind people like you will help me to speed up my learning curve.

You are encouraged to learn, so feel free to ask what you don't know. I recommend you to use AWS S3, not other parts of AWS. AWS S3 is an object storage solution. It can function like dropbox to you, but one good thing is it can serve static webpages from it too. Feel free to read up from How to Host a Website on S3 Without Getting Lost in the Sea.

The support team at Resellerclub.com asked me to verify the server performance using https://gtmetrix.com
Is this correct approach?

You can use this for testing the performance of your website's pages. It does not only test your hosting in particular. Services like this, similar to Google's Pagespeed benchmark your website's pages. The time to takes to download the resources, the time your pages render, whether resources are compressed and so forth. Some part of the benchmark is dependent on your hosting plan configuration of the web server and network. Some parts are totally dependent on your web design on how it renders on the the browser. GTmetrix by default access your webpages from Vancouver, Canada. You can login (not sure free or not) and change where you would like to test your site from. The further your client(browser) is from the server, the higher the latency and other network conditions. It uses Google's Lighthouse too, so you can go to https://developers.google.com/web/tools/lighthouse to use your own installation (if you know how) to test your own website accessing from your own house/office/country.

 

[GoldmanSach]

woah, do you happen to purchase the cheapest variant

static sites may host rich media, check disk storage. 3 domains/subdomains, email addresses, and cPanel do consume resources.

anyway, you need to ping the shared hosting to see where is the server’s location vis a vis your ftp location.

if latency is normal, then check if your vps or basic shared plans lack the ram, else is really the static websites not build properly (ie WP template has bugs)