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
Guess that's a compromise. LET hosted on a high end server to accommodate hell lot of low end talks.
Memory for these plans are twice the number of CPUs. 16G for 8 CPU, 32 for 16, and so on.
Actual memory usage would be a nice stat to have, personally I have no clue what that's like for a VanillaForum.
But this 2:1 ratio (2G RAM for 1 CPU) is very popular for web hosting.
Generally for a given application you size the memory according to the number of CPUs, and the number of CPUs on the per-core performance of the application.
For example, if your app uses 4G memory per CPU (100% utilization) and can serve 1000 concurrent users, then you need to provision a 4 CPU / 16G RAM server to handle 4000 concurrent users.
Edited: basic math 🫣
Pavin.
Thanks a lot for this. I will keep this in mind.
I was hoping for a fleet of LE servers with some cool clustering, but this is so vanilla.
No need to involucrate me, I'll show myself out 🙃
Pavin.
LOL! Woulda been cool, but, as jb said, since they temporarily upgrade cores during special periods, I guess something like Premium DO is the best option.
When can we expect 404talk?
(404talk.com still available as we speak - grab it now before someone else does!)
Nope. Was asking most of this for curiosity. Do run a simple non-hosting related forum on shared hosting. Don't think it will ever have to be upscaled to the level of LET.
it's on a galaxy near pluto in a lot of cloud
Is there a Plutonian provider here providing the service and accepting payments in PayPal.
How can anyone research which web hosting provider any particular website is using? (Asking generally, not regarding any specific site.)
You generally have to ask these days, since most are using a reverse proxy, it will show the reverse proxy host. You can sometimes find clues in DNS history.
Where can you review the DNS history?
And if they aren't using a reverse proxy, how can you then see why their host is?
https://dnsdumpster.com
It used to be that the nameservers gave it away, but those can be whitelabelled too.
Was the NS the main giveaway, historically? It was common already long ago to use third-party NS, unaffiliated with your hosting provider.
I would have thought that traceroute, nslookup and WHOIS of IP addresses were, at least at one time, better indicators.
.
.
with autoboot
I really needed that laugh.. whoever is responsible for that beauty deserves an actual tonguekiss
Petition to move from digital ocean to oplink
My guess is that in an effort to try and stay professional/unbiased they don't use LET providers, otherwise there could be conflicts of interest.
high tech, low life.
Nope, premium Low End Talk.
Good question actually. Most people seem to think in CPU only (and preferredly Threadripper or Ryzen) but actually oftentimes memory is more important that is, weaker processor but lots of memory - is better than - stronger CPUs (or more cores) and less memory. Largely depends on application type (for web services it's largely as I said) as well as on architecture (e.g. multithreading vs coroutines or async/await).
Thanks for that input. But, I guess the CPU cores actually do matter when it comes to concurrency. For eg:- running Apache benchmark while consequentially increasing the concurrency, I have seen my test system's core usage hit 100% while the RAM's usually at 50%.
But, you are right. It all comes down to the application type and it's demands.
It's unfortunate that there isn't a single simple tool that informs people of recommended VPS specs for atleast some popular applications. There are blogs talking about it but they always talk in extremes.
how about load balancing across multiple top hosts lowend providers?