Live data from Hacker News

Hetzner Price Adjustment

docs.hetzner.com

241–250 of 312 posts

Re: Hetzner Price Adjustment

#241

Earlier quoted context omitted.

My R730 averages at around 100-200W with two 12-core CPUs, 128GB RAM and 8 12-14TB drives. Servers are more efficient than people give them credit for. It's close to a gaming PC in electricity usage. Certainly not 60W (or what people can get with NUCs and RPi clusters) but for the power I get, it's very much worth it.

If we average that out to 150W, that'd be 1250-1300Kwh/yr. Current Bay Area electric prices means that'd cost in excess of $500/yr to run. I'd say that servers are not more efficient than people are giving them credit for...

And at Seattle prices, that's $150, so I don't exactly see what the purpose of your anecdote is other than to mention how pricy your electricity is.

Edit: But even at $500, that's pretty much the price of a low-spec VPS or VM per year. So for a fairly low initial price to buy the server, you're getting far more performance for the same price you'd pay someone else to use theirs.

Re: Hetzner Price Adjustment

#242

Earlier quoted context omitted.

Highly disagree. Said in another comment, my server averages around 100-200W. It's equivalent to running a gaming PC. The reason there's a lot of cheap servers on eBay is that their support contracts expired and are absolutely ridiculously expensive to maintain for a big company. It's easier to get a new server (which will likely have performance and efficiency upgrades as well) than to continue maintaining a server…

A fan replacement and activating quiet mode on the server (usually a bios option or sometimes a hardware tweak) helps with the noise a ton at the cost of slightly less thermal ceiling.

That's why I said "by default". I have done the IPMI toggles needed to lower the fans but it's still not quiet by any stretch of the imagination.

Re: Hetzner Price Adjustment

#244
post #47

Its fascinating that electricity is really the biggest cost of computing. I put a server in my house which was supposed to save money, but actually the electricity cost was something I hadn't thought of and really adds up. esp when the AC has to run to remove the extra heat. So much old hardware out there but its going to get junked as new hw is more efficient.

As a rule of thumb, 1W costs about $1 per year.

Re: Hetzner Price Adjustment

#245
post #177

Earlier quoted context omitted.

I would say the big mistake was to only use pipelines to Russia for gas. They should have built LNG terminals as redundancy. And maybe push for another pipeline to Algeria either through France and Spain or Austria/Switzerland and Italy. They basically left the door open to abusive trade partners by not building any redundancy on such a critical resource. And it is not just electricity Germany needs gas for as it is…

I’m just going to point out that Trump said this exact thing and was ridiculed for it: https://www.washingtonpost.com/world/2018/09/25/trump-accuse...

I'm quite sick of that "reverse ad-hominen" attack on politics of discrediting ideas just because somebody stupid agrees with them.

Re: Hetzner Price Adjustment

#246
post #4

While the pricing is interesting, I think the fact it is driven by the input costs of electricity is even more telling. The fact that electricity costs is the dominating (edit, correction: marginal ) cost of cloud computing basically means that: 1) We're up against physical barriers to lowering the cost (and ecological impact) of our clouds. 2) Optimizing for less compute will pay multi-fold dividends over the next f…

It's been like that for a long time (not only for cloud computing, but any big IT infra). Power is everything. To a point, that you measure datacenters in megawatts, when you bring new HW to datacenters you most often talk about improvement in compute per watt, not absolute compute.

It's not only because of the cost of power, but also because power is something that's hard to increase in an existing datacenter - you can put in new servers easily, but putting new power lines and upgrading all the infrastructure is usually just too expensive.

Re: Hetzner Price Adjustment

#247

Earlier quoted context omitted.

I wonder if latency allows this

It depends on your requirements. Serving user-facing content is likely to be a task best suited to data centers geographically close to the user base. There are plenty of non-user-facing tasks that could be offloaded to another location. As an example, the product I am developing spends most of its time calling third party APIs to do bulk data extraction and processing. That can happen anywhere. After data processing…

With cloud egress cost though, moving the data to a low-cost region could be expensive itself. I wonder when it would pay off.

Re: Hetzner Price Adjustment

#249
post #192

Earlier quoted context omitted.

> They wanted to be super green so lets go with sun and wind This is a complete mischaracterization of why Germany invested so much into renewables. Like... 20 years ago? Germany saw the writing on the wall when it comes to how they were getting their energy which at the time was mostly coming from coal. They didn't want to go Nuclear for whatever reason so that left renewables and natural gas which wasn't preferred…

I don't say that that being super green is bad, the problem was to rely almost solely on natural gas (from those we don't name) to stabilize the grid. Well, and now using that bad coal to fix that... I don't know, that also doesn't seem to be very wise, although I know, its only the consequence of the first mistake, but it completely negates what they tried to achieve.

Every single country relies on either natural gas or hydro power to stabilize the grid. What alternative did you expect?

If there is a problem, it's either on the share of the generation that they decided to let for those "stabilizing" sources (it is much larger than the short term instabilities on their grid) or on the lack of diversity of their supply.

Re: Hetzner Price Adjustment

#250

Earlier quoted context omitted.

The rush out of nuclear fission power still shocks me. It's a miracle fuel. It really is wonderful. It solves climate change almost by itself, or it would have, had Chernobyl and Fukushima not been reported so irresponsibly.

It would be miracle fuel if any country in the world would have solved the waste problem by now. Finland is close, but still not done.

Research what France does. They recycle the waste because it is not waste, it just needs processing and can be used again. https://www.iaea.org/newscenter/news/frances-efficiency-in-t...
Post reply on HN