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
I'd say it depends what you are using it for. But usually, as any computer, the slower component will be the storage. Especially if it's HDD.
I mean which is the limitation in order to upgrade to another higher specification vps kvm .
in my experience, it is CPU
I had issues with resource usage and upgraded the cpu and it works fine
There is no answer to this question. It is like asking "what stops a car going fast" without saying what type of car it is.
A Wordpress based website.
You are best to look at your resource usage and see where you are getting bottlenecked, Normally for a site it could be the drive depending on the vps you have, otherwise it may be your setup for the site
Storage probably. Most of the time if you upgrade/downgrade your disk space, you have to resize the file system yourself.
CPU and RAM does not have much impact in the process, unless your applications have some special configuration for that.
For wordpress, cpu is usually the bottleneck.
not true. for wordpress in 99% of cases the bottleneck is the idiot in front of the monitor calling himself "webmaster" while installing plugin number 324...
Sad but true
I am completely agree with you . My website has just 15-20 requests and there are websites with 300 or more requests . I like wordpress . there are lot of limitations but ........i like it
You are playing a dangerous game. Bottleneck is always going to be there. And that term in computers came to life few years ago when some nerds had nothing else to talk about.
You will pull your hair for nothing.
Bottleneck is budget
The difficult part will be disk (often this involves manual steps). In terms of scalability I would say most likely cpu. Disk isn't too expensive and ram isn't too bad, but there's physical limitations on cpu.
The bottleneck is usually the programmer.
If you cache your pages and serve static files using nginx or varnish, or apache in mpm events your only bottleneck can be ram (you will need some so that the kernel can cache your files in RAM, avoiding the need for high i/o) or CPU (for tls negotiation as serving static files doesn't need much CPU) or... network (if you have enough CPU to saturate the port) you'd need a LOT of traffic though.
this is lowendtalk we must be cost effective !
I use cloudflare + nginx and the vast majority part of the website is static.
it is me
but i work with 15-20 request per page so .....but i must recognize it i am the bottleneck anyway , you are right
CPU ? I am using cloudflare+nginx +cache for the static elements in order to save cpu
You are right the bottleneck is going to appear in the vast majority of cases , but i was curious about which is the general opinion respect it . Unfortunately , i am not so nerd to solve my all issues so I upgrade 1 step before hand in order to have more resources . I am not cost effective but i must pay the lack of knowledge so it it fair for me
The user.
@Chievo seems like you need to study what's going on with your server to know what is the bottleneck, rather than just buy more hardware: how is cpu usage? iowait? ram? swap? etc...
Do you really use full page cache with nginx serving directly the pages with a config similar to that: https://www.keycdn.com/support/wordpress-cache-enabler-plugin#nginx ? Can most of your pages be cached and generated once for several days?
If so, unless there is a LOT of traffic and TLS negociations take too much CPU or you don't have enough RAM, there should be no bottleneck.
You asked which is the limitation with upgrades, not which is most likely causing you performance problems.
also keep in mind overall throughput may oftentimes be limited by php selecter ok
Yeah, a good php selecter is probably the answer to most bottle and neck problems.
OK: my knowledge and experience level is low enough i won't even know if this question is nonsense, hoping for a patient explanation, if it is at all possible:
Suppose the VPS is used for WordPress websites, that are decently optimized (using caching plugins and server's cache - LiteSpeed, or nginx).
What resource "ratio" is most likely to be "optimal"?
That is: I suppose 1 vCPU core and 1 GB of RAM will be used with cPanel when it is being used. So is 2 vCPU and 2 GB RAM an OK ratio,
or is it better to look for 2 vCPU and 4 GB RAM (or the other way round)?
How is it best "scaled" for more resources? Same increase in RAM and CPU, or different ratios?
With an average of 1000 daily visitors per website, how many such websites can run nicely on such a VPS (I guess this must also depend on how many VPSs with what kinds of websites are created on a physical server, especially for I/O and IOPS), is there some "general" estimation?
Or is the only way to go testing and seeing? I'm sure that's the most exact way, but would like some "rough figure" for price calculation.
P.S. Of course I'll disregard your signature and also get at least one VPS that is left idle, so I could comply with the LET rules.
I don't use shared hosting enough to know for sure, but I am convinced that when you have 1 GB of physical ram on cpanel, you don't really get to use 1 GB of ram. Though 1 GB is not a lot of ram, in reality, it actually is quite a lot of ram and it's not feasible for anyone to give shared accounts their own dedicated ram. Maybe the vswap is very swappy, maybe ram is cycled/shared, maybe they kill your processes, maybe you're not expected to use things like memcached/redis etc. And it's very very VERY easy to get throttled by cloudlinux just using the wordpress backend. See, vps is a shared environment, maybe you have 100-200 users on a server max, possibly less. While you could be looking at 1000+ shared accounts on just one server alone if your provider sold reseller plans. You're comparing apples to oranges, because your 1 core and 1 GB of ram is more shared than you think.
It's hard to say, but it's best to have at least 2 threads for wordpress sites in my opinion. Most likely for small sites, you don't need a lot of ram. It's only when your ram grows big and you want to cache your database where you need more ram.
With a 1 e3 core 1 GB ram vps, I wouldn't be surprised if you can run 30-50+ of such sites. Maybe even 100 of these sites, but with a plan like that, you probably won't have enough disk space for all those sites. Think about it this way, if your visitors are mostly spread across let's say 12 hours a day, that's 43200 seconds. If you have 100 of these 1k daily view sites, thats 100k views, so just around 2.5 connections per second. Of course, maybe you have peak hours so it's not that simple but webhosting honestly doesn't take a lot of resources. The wordpress backend, or other things like woocommerce however are resource hogs and you can't work on several different sites' backend at the same time.
Edit: I take back what I said, thinking about it, I think it may be best to have at least 50mb of ram per wordpress site or it's quite likely something will crash/break. But it doesn't change the essence of what I wrote before this edit. You probably could squeeze 20 sites that have 5k daily views into a 1 gb ram vps by having some swap, but you may get some errors occasionally. But just because you can, doesn't mean you should.
tl;dr managing websites sucks
MySQL.