Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Shared hosting or VPS, Actual benchmarks / comparison
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.

Shared hosting or VPS, Actual benchmarks / comparison

Good day :) ,

There are many websites out there that give reviews on top shared hosts, where they break down various aspects, such as speed, reliability, quality of tech support, etc.

My concern is that a lot of them give preferential treatment to some of the big names in the business, with a conflict of interest since such reviewers may get a commission for referring folks to the various web hosts they list as "the best."

I'm wondering, from those who are web developers... (such as those that work with Wordpress / Ecommerce platforms that perform better with top hardware (NVME storage/AMD Ryzen or similar), which hosts have you found to be the fastest to USA audience? (in terms of disk I/o, Network performance (mib/s)
In other words, many hosts put caps on disk I/o and bandwith mib/s and understandably so.

However, I'm wondering which hosts seem to be the most generous in regards to these limits?

...and most important of all, "real-life" speed / when one does benchmark tests with tools like GTMetrix etc?

The eCommerce platform I'm using is not Wordpress, so unfortunately I don't think it can take advantage of Litespeed caching etc.

If you can comment the web hosts names and the particular plan/package you use/have used, that'd be great. Also I'm open to consider VPS or shared. I'm guessing depending on the host, that sometimes shared could actually give one more available resources than VPS, depending on their generosity, etc.... although I recognize there's Acceptable Use policies etc.

Thanks for taking the time to share :)

Comments

  • bruh21bruh21 Member, Host Rep

    We offer barebones shared hosting which is admittedly more cost-optimized than speed optimized (not to say it’s slow). I’ve had good experiences with racknerd performance (their IO is generous) and nexusbytes also has very generous IO with their plans (although their Germany location has had downtime issues).

    @MikePT also has good hosting.

    Thanked by 1MikePT
  • MikePTMikePT Moderator, Patron Provider, Veteran

    @bruh21 said:
    We offer barebones shared hosting which is admittedly more cost-optimized than speed optimized (not to say it’s slow). I’ve had good experiences with racknerd performance (their IO is generous) and nexusbytes also has very generous IO with their plans (although their Germany location has had downtime issues).

    @MikePT also has good hosting.

    Thank you! :)

  • CoastHostingCoastHosting Member, Host Rep

    that is quite an open ended question. WordPress can run on very minimal specs providing the pIugins used are good or you have optimized your site well. Having high I/O is great for badly designed sites. We offer 10Mbps I/O with 2GB memory and 2vCPU LVE limits on our wordpress plans. Most wordpress providers only offer 2-5Mbps. So it does depend your purpose.

  • fast I/O on SSD or NVMe, faster CPU, less usage of RAM by web host's customers and congestion-free multi-homed network (2 or more upstreams connected) can do wonders. Even MySQL connections, indexing tables, use of Litespeed server than Apache, can benefit the speed on shared hosting which can be as much as renting a VPS (and managing all by yourself).

    Some hosts offer 80MB/s I/O on SSD, some offer 5MB/s on NVMe. Now that's the thing of comparison if you are going to host large wordpress sites. Also the number of users on one single machine also will affect the website performance. For example, Host A using E3 (8 core) machine may have 400 cPanel accounts, while Host B using latest Ryzen (24 core) can place 2000 cpanel accounts all using the same disk I/O, my sql, etc. Here performance of hosting on Ryzen can be sometimes degraded than hosting on E3 during peak hours where all sites receive much of the traffic. So again this is the point of comparison.

  • @techmon said: However, I'm wondering which hosts seem to be the most generous in regards to these limits?

    ExtraVM has 80 MB/s I/O limit which is the most generous I've seen. Limits of 5-10 MB/s used to be relatively generous in a time when godaddy was still giving 0.5-1 MB/s, but nowadays even godaddy gives 10 MB/s limits (just googled this).

    @techmon said: ...and most important of all, "real-life" speed / when one does benchmark tests with tools like GTMetrix etc?

    Loading the front page of your ecommerce site will be really fast. Things may load much slower for customers that are logged in and trying to buy things, or when you are using the backend to add products etc. For a serious ecommerce website that is going to make money, there's really no reason to consider shared hosting.

  • paijrutpaijrut Member
    edited April 2022

    Every website is unique on how much resources needed. Some shared hosts are actually faster compared to vps when configured poorly. You have to try it yourself for better insigth.
    For shared host I'm using racknerd @dustinc on US and it's great for my purpose (litespeed, cpanel, and mailing), better than when I'm using vultr vps for this particular website. For other purpose I'm using vps on my local provider (local audience only). Its all depends on your script, traffic, and other requirements you had..

    Edit : also depends on your sysadmin skill if you opt to use unmanaged vps/dedicated

    Thanked by 2ElonBezos bruh21
  • ZerpyZerpy Member

    @NoComment said:
    but nowadays even godaddy gives 10 MB/s limits (just googled this).

    For new accounts only. "legacy" users (despite being on their "current" plans), are limited to 1MB/s - I had the honor of migrating multiple customers from them over the past week, and the 1MB/s limit is absolutely horrible when you have to move tons of data 😆

Sign In or Register to comment.