Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Hetzner launches three new dedicated servers

hetzner.com

Post reply on HN
Print view View original

231–240 of 320 posts

Re: Hetzner launches three new dedicated servers

#231

Post by izacus »

Symbiote wrote: post #40
danwee wrote: post #31

Earlier quoted context omitted.

> Then you contact support, appoint disk change, you first deactivate disk on raid (save geometry etc), they replace disk and then you rebuild raid in new disk. That's it. I imagine this would take time, right? Like not 5 minutes, but maybe 3 hours top? So, if I pretend to run a saas (that shouldn't be down more than 1h/day), then renting only 1 dedicated server could be qualified as "risky"?

I'm not sure what "deactivating RAID" means. They will all be hot-swap disks. You remove the old disk and slide in the new one (or in this case, tell them to do it). The RAID system rebuilds the array in the background over the next few hours. During that time you will lose data if it's RAID 5 and another disk fails.

> I'm not sure what "deactivating RAID" means.

mdadm --manage --remove

so your machine doesn't have a fit when the disk is detached. Or equivalent.

Re: Hetzner launches three new dedicated servers

#232

Post by abujazar »

whalesalad wrote: post #215
abujazar wrote: post #131

I've been using Hetzner servers for ~15 years with multiple clients and employers, and always been disappointed with other providers compared to what Hetzner delivers. OVH with their frequent network level outages, the 2021 fire and so on. DigitalOcean with their way too frequent and long lasting maintenance windows. And AWS/GCP/Azure with their obscene pricing, ridiculous SLA and occasional hour-lasting outages. One…

Just wish they had more of a US presence. Latency is a killer for me.

They do have two US locations, but only for «cloud» products, no bare metal servers.

Re: Hetzner launches three new dedicated servers

#233

Post by indymike »

outime wrote: post #164
abujazar wrote: post #131

I've been using Hetzner servers for ~15 years with multiple clients and employers, and always been disappointed with other providers compared to what Hetzner delivers. OVH with their frequent network level outages, the 2021 fire and so on. DigitalOcean with their way too frequent and long lasting maintenance windows. And AWS/GCP/Azure with their obscene pricing, ridiculous SLA and occasional hour-lasting outages. One…

While I like Hetzner a lot and can share your recommendations, I just don't see how it compares to full-blown cloud providers like AWS, GCP or Azure. It's a common misconception to put them at the same level when the offering is completely different. Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone. If someone can run their full workload in e.g. Hetzner without much hassle then they…

> Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone.

Many companies and people do host loads that would be better served on dedicated hardware on EC2 because "cloud".

> Hetzner without much hassle then they shouldn't be using any of the other cloud platforms in the first place as they'd be definitely overpaying.

The ability to provision, de-provision, clone, load balance and manage without talking to people, waiting for hardware or really even having to understand in detail what is going on (yes this is bad, but still... ) is one of the big reasons cloud is popular. Many dedicated hosts have gotten a lot better in this area.

Re: Hetzner launches three new dedicated servers

#234

Post by daydream »

mhaberl wrote: post #218
outime wrote: post #164

Earlier quoted context omitted.

While I like Hetzner a lot and can share your recommendations, I just don't see how it compares to full-blown cloud providers like AWS, GCP or Azure. It's a common misconception to put them at the same level when the offering is completely different. Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone. If someone can run their full workload in e.g. Hetzner without much hassle then they…

> Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone. Most of my clients do just that - just EC2 on AWS. Ofcourse, my experience may not represent average case, but it is certainly not "nobody". I believe that most do it because AWS/Azure is the "safe option". Choosing AWS/Azure is the modern version of "Nobody ever gets fired for buying IBM". -- I just recently tried Hertzner myself…

Lots of companies do this even some big ones. Mitigating vendor lock-in is a big reason. Using what’s effectively simple VMs makes it much easier to pick up and go elsewhere.

Not all businesses decide that’s a risk worth mitigating, but some do.

Re: Hetzner launches three new dedicated servers

#235

Post by fatnoah »

bluedino wrote: post #195
outime wrote: post #164

Earlier quoted context omitted.

While I like Hetzner a lot and can share your recommendations, I just don't see how it compares to full-blown cloud providers like AWS, GCP or Azure. It's a common misconception to put them at the same level when the offering is completely different. Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone. If someone can run their full workload in e.g. Hetzner without much hassle then they…

> Nobody seriously uses AWS/GCP/Azure to have a couple VMs or dedicated servers alone. I think you're seriously underestimating the amount of cloud customers that do a simple lift and shift.

> I think you're seriously underestimating the amount of cloud customers that do a simple lift and shift.

I've done exactly that at a previous startup. Granted, it was 10 years ago, but going from racked infra to AWS ended up being half the cost for what was effectively twice the infra (we built out full geo-redundancy at the same time).

Re: Hetzner launches three new dedicated servers

#236

Post by nmjohn »

izacus wrote: post #229
nailer wrote: post #128

Earlier quoted context omitted.

The expectation with AWS and GCP is that the constant intermittent failures are shouldn’t be visible to end users. * So filing tickets is a big difference in user requirements. * Though GCP (back when it was just AppEngine) wasn’t always this way and as GAE users inside Google we had to write our own code for what we expected to fail, retries, backoffs etc.

Yeah, turns out that if you write code for constant intermittent failures, it'll run well for rare intermittent failures of Hetzner machines too.

With AWS - failure means the instance is automatically retired, and your ASG causes a new instance to automatically be created and put in service without you having to do anything.

With hetzner - failure means your monitoring detected disk failure, sent you a pagerduty alert, which you then have to check the alert, figure out what has failed, and send in a support ticket to get the disk replaced. This will take a couple hours, after which you have to rebuild your RAID array, and hope no more disks fail. All the while operating with degraded performance.

(Don't get me wrong, hetzner is _great_, I've used them for years and highly recommend for numerous scenarios - but the idea that their failure and reliability is anything like "the cloud" is fanciful)

Re: Hetzner launches three new dedicated servers

#237

Post by efxhoy »

jhoelzel wrote: post #37

I have been working in the industry since 20 years and i have not a single bad word to loose about Hetzner. Their Service has always been impeccable and their servers just run. I have been running k8s clusters on hetzner for quite some time now and the flexibility for the price is exactly what i expect from a hoster! Now with this addition Hetzner closes another gap that made projects spends thousands more on enterpr…

How did you setup the k8s cluster? Is it running on bare metal or VMs?

Re: Hetzner launches three new dedicated servers

#238

Post by karussell »

pornel wrote: post #221
karussell wrote: post #81

I have a question. These newer Intel machines all have performance and energy efficient cores. This makes sense as they are designed for the Desktop, but is this meaningful for server-side applications? Recent Linux kernels finally support these CPUs (do they have full support?), but if you host a service where you want predictable (and fast) response times why you use the mix of both cores? Or would you just turn of…

I don't expect the efficiency cores to be slow enough to have noticeable effect on response times. Rather they'll just have smaller throughput, and other cores will pick up the rest when the server is loaded. I'm assuming you don't shoot yourself in the foot by running strictly single-threaded workflow explicitly pinned to the efficiency cores.

We have tried it for the older generation and there is definitely a different response time. Why should it be different? Surely our workload is more computation intense than others but I'm pretty sure that you'll get a difference for other workloads too.

> running strictly single-threaded workflow explicitly pinned to the efficiency cores

Those cores are slower than e.g. the cores from the (Desktop) AMD CPU we tested at the same time (offered from Hetzner). So it is rather expensive and inefficient to use Intel (Desktop) CPUs for server-side applications as we can only use their performance cores.

Re: Hetzner launches three new dedicated servers

#239

Post by usrusr »

lloydatkinson wrote: post #180
adql wrote: post #124

Earlier quoted context omitted.

Maybe he detected 0 RPM without looking at temperature.... some controllers will just stop fans when there is enough cooling. We had a bunch of racks and fan failures gotta be one of rarest one. Even on my personal junk I had zero actual failures , just one getting noisy > Was still worth it because it's cheap, but they're cheap because it's not new and reliable hardware. Not really different than cloud, they aren't…

Are controllers that throttle down to zero common in the data centre space though? I don't think I've ever heard a quiet or silent server, except for the ones tech people make themselves at home.

Hetzner is known to do their own thing instead of blindly following datacenter conventions. E.g. having the physical architecture (how their buildings are designed) so optimized for convective flow that they get sufficient air flow without extra fans most hours of the year and only need active "heat pump" style cooling in rare exceptions. They do have board-level fans, but it it would be almost out-of-character if those did not throttle to zero whenever possible.

Big german language report: https://www.golem.de/news/besuch-im-rechenzentrum-so-betreib...

Re: Hetzner launches three new dedicated servers

#240

Post by derefr »

Question for the audience: if Hetzner's highest-specced bare-metal servers would work great for my use-case, except that my customers are mostly in America and my service is realtime-ish enough that I need a low-latency RTT to them, then what's the next best option? What's the "Hetzner of North America"? (I know Hetzner has US datacenters, but they don't do any dedicated hosting there AFAIK, only cloud hosting.)

Personally, so far, the best near-equivalent provider I've found that actually offers well-specced machines in North America, is OVH, with their HGR line and their Montreal DC. Are there any other contenders?

And if not, why not? what's so hard about getting into the high-spec dedicated hosting space in the US specifically? Import duties on parts, maybe? (I've found plenty of low-spec bare-metal providers in the US, and plenty of high-spec cloud VM hosting providers in the US, and plenty of high-spec bare-metal providers outside the US; but so far, no other high-spec bare-metal providers in the US.)

Post reply on HN