Anybody have tried Google Cloud web hosting service?

[HostXNow]

You want me to set up an optimise a wordpress site just for the sake of proving something to you?

I don't want you to do anything, but if you're trying to back up what're saying or try to get your point across better then maybe it's in your interest to do so. Even if you put a default WordPress install using the default theme on a sub-domain hosted on a setup you think loads a website faster, then that'd do for the test. I'm sure I could get it to load more quickly with the setup/optimisation that I do. If not, then I stand corrected, but at least I'm willing to back up what I'm saying. You're just saying such and such is such and such and mostly giving apples to oranges comparisons.

It's probably best not to say anything if you aren't willing to back up what you're saying. Posting lots of links to static HTML sites is not saying Apache is faster than LiteSpeed or dismissing what I was getting at in my initial post.

You can't expect me to cover the topic 100% with a paragraph or two just as I wouldn't expect you to cover the topic 100% with only a paragraph or two. It's a very complex topic.

I gave information in my initial post that will help. I never said it was the bee all and end all haha. If someone would like further information on how to make a site faster, then they are free to ask for more information.

Also, this was about the webserver, but then the discussion changed to how fast the website loads depending on the code.

Nothing you posted backed up what you were saying. Saying some static websites nearly loaded as fast as a dynamic site isn't saying Apache is faster than LiteSpeed. If it were a static website on Apache and Litespeed, then LiteSpeed would have loaded the content even faster than it did with dynamic content.

Coding aside, out of the box LiteSpeed with LScache will load a website faster without having to tweak Apache or Nginx to death. If you prefer to to go about it the long way for each site to get the same or similar result, then that's up to you, but that takes a lot more time and is more for someone who knows how to do such tweaking.

A lot of individuals or businesses haven't got time to do that or even know how to do it. When hosting on a setup with LiteSpeed/LSCache it's already optimised, and it's possible to tweak things even further if needed. Each to their own.

LScache is very advanced. LiteSpeed is faster than Apache and Nginx by default, even faster if you tweak LiteSpeed. Nginx plugins for cPanel mostly proxy only, so nowhere near as good as LiteSpeed. LiteSpeed is a drop-in-replacement and makes websites faster out of the box regardless of how poor or good the coding is. LiteSpeed also fully supports .htacces, whereas not the case with Nginx.

My point to you was to say LiteSpeed is faster than Apache out of the box i.e no tweaking necessary.

Maybe, so are a lot of other options.

But not Apache vs what I put. The things I said that make a difference, and they do make a difference. You said they were too generic. I also said NVMe helps and with the test I provided, the site is hosted on SSD, not NVMe. So likely even faster if hosted on NVMe.

It is a different thing, and usually much more important.

I agree with this, and at any point did I said it wasn't important. Nor did I say only the hardware/network/software plays a role. You're right; the coding is usually much more important. But from the start, I've said a site with the same content would load faster on LiteSpeed than Apache. LiteSpeed would also handle concurrent connections a lot better than Apache would, i.e. handles massive traffic much better without server running out of resources, or worse, crashing altogether. There are many tests which show Litespeed handles traffic better than Nginx, too.

If people want a faster site they do not care whether it is achieved by optimising the design, or optimising database queries, or changing the web server - but for most sites the problem is the design, the code or the database (most often in that order).

Again, I agree with that, as mentioned in my previous comment. I never said that wasn't true. I was merely saying that out of the box LiteSpeed is faster than Apache and that it depends on storage being used as well as the resources limits that a provider offers to their customers, i.e. if a cron is running in WordPress and the account has 1MB/s Disk IO it would take a lot longer for it to complete than on an account that has 50MB/s. When a resource is being throttled due to a cron running, it can slow down the website for other visitors visiting the site at the same time.

As I said, one could write many pages on how to get the site to load as fast as possible, but the ones I mentioned in my initial post can make a huge difference regardless of how good or bad the coding is in the website.

I'm well aware the website should be optimised first, and only then put it on the best network/hardware/software and optimise further.

 

[gpietersz]

Posting lots of links to static HTML sites is not saying Apache is faster than LiteSpeed or dismissing what I was getting at in my initial post.

Nor is posting links to your sites proving anything either. I also provided you with a dynamic site for comparison that is faster running on Nginx, without caching, than your site is on Litespeed.

What you actually originally said, and all you said in your original post, was:

Yes, try to use a provider who uses CloudLinux, LiteSpeed and also the newer and faster NVMe drives if possible. Your websites should load much quicker then.

As you have now said

the coding is usually much more important.

Which is it? If the coding is much more important, changing the webserver will make an insignificant difference. You are not going to get most websites to "load much quicker" by changing the webserver or using a different cache.

You have now agreed with what I said in my original post so i do not know what you are arguing over.

 

[HostXNow]

@gpietersz

It is what I said. What I said would help make the site faster if not already using said software/hardware.

You are then free also to mention that optimising the website can help make the website faster. I could have said that, but I instead let several members add different things, rather than me trying to cover everything in one go.

You didn't need to dismiss what I said. You only had to mention what else you think could help make a website load faster.

You quoted me first and tried to dismiss what I said. Not the other way around.

Which is it? If the coding is much more important, changing the webserver will make an insignificant difference. You are not going to get most websites to "load much quicker" by changing the webserver or using a different cache.

You have now agreed with what I said in my original post so i do not know what you are arguing over.

That is wrong. I did not agree with that.

I said coding aside, switching from Apache to LiteSpeed, using NVMe instead of SSD will likely make the site load faster. Optimise the code, and it will load even quicker. But it will still load faster after switching, even if the code is the same.

 

[AllUpHere]

Just when we thought a thread about hosting couldn't get any more tedious.

 

[HostXNow]

Just when we thought a thread about hosting couldn't get any more tedious.

Imagine discussing all that in a packed pub on a Friday night.

 

[gpietersz]

Just when we thought a thread about hosting couldn't get any more tedious.

Really? You do not know the right people (or possibly the wrong people).

 

[TopSpek]

I've been quite happy with a2hosting. I had a reseller package for about £30 per month, and the vast majority of the time my site seemed to be somewhere around 96 to 100 at Google Page Insights for desktop, and 84 to 90 for mobile. My site has absolutely no optimisations whatsoever - it's just raw PHP/MySQL built according to my own MVC framework which I've developed over the years. Oh, there's a hint of JavaScript there as well, but I'm going to dump this.

I recently downgraded to a straight shared hosting package for about £12 because that's really all I needed; I was never going to become a webhost because, as HostXNow has clearly demonstrated in this thread - it's a full time job and requires lots of expertise. Anyway, after I returned to shared hosting, my Google Page Insights rating dropped to between 95 and 98 for desktop, which is fair enough - I'm not going to lose any sleep over that because I can intuitively appreciate that there would be good reasons for why this occurred.

I don't get too hung up over the server hosting details too much because that's not my job. It's much more efficient for a software development system to leave things like that to the specialists, but obviously I need to know enough about it to enable me to do my job; for example, setting up websites on my local machine in an environment which mirrors the remote server closely enough for accurate development and testing before uploading to the remote server.

So, IMHO, I think it's best to just find a trusted recommendation (perhaps the one I've provided here) and get on with your life. You don't need to delve too deeply into the details. And after reading HostXNow's messages in this thread - well, he sounds like the second hosting expert I've found on UKBF in as many days.

 

[enjoylife]

Updated: we have transferred two of our websites to Guru in Nov. So far so good, we can see more traffic after using them. Recommended.

 

[DaliaBall]

I haven't tried yet because of cost of service.at this moment, I don't have enough money to invest on google cloud service.so, I think cloud hosting will be great choice for high traffic website owner.