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

Hetzner Introduces ARM64 Cloud Servers

hetzner.com

Post reply on HN
Print view View original

281–290 of 342 posts

Re: Hetzner Introduces ARM64 Cloud Servers

#281

Post by viraptor »

simplotek wrote: post #268
MagicMoonlight wrote: post #217

Earlier quoted context omitted.

Isn’t this entire site run off a laptop in someone’s shed? And yet it’s got basically perfect uptime. Whereas the impenetrable cloud goes down every week and costs billions a year.

> And yet it’s got basically perfect uptime. Some monitors indicate that hackernews has a 99.84% uptime in the past 30 days. Hardly "perfect". https://hacker-news.statuspal.io/

And that still doesn't include the very short term soft-failures where HN tells you "something happened, try again" - those happen every week or so.

Re: Hetzner Introduces ARM64 Cloud Servers

#282

Post by jacooper »

usr1106 wrote: post #265
solarkraft wrote: post #51

Earlier quoted context omitted.

Oracle's free tier is cool. I've let my free instance expire and so far haven't noticed them doing anything weird with my credit card info. Also haven't been sued yet. But let's see.

Oracle did not accept my credit card. The same one AWS has been charging successfully for years. But I don't really want to do business with Oracle anyway, maybe it's better so.

Can confirm, they are very picky about the credit cards they accept.

Re: Hetzner Introduces ARM64 Cloud Servers

#283

Post by martinald »

glutamate wrote: post #279
simplotek wrote: post #266

Earlier quoted context omitted.

> So what's the point of all that? Seriously, one server with 16 cores and the memory bandwidth and storage speed of SSD drives should be able to serve many thousands of simultaneous requests and millions of users per month. I can't help but feel that the cloud infrastructure and microservice movement of the 2010s was.. a scam. You're just missing the whole point of why you need n>1 instances. You're focusing on what…

A simple single node is often more reliable than a complex fault-tolerant setup. Fewer moving parts mean fewer broken pieces. Similar with security. A simple monolith is more secure than isolated front- and backend.

Agreed entirely for most. I have seen so many insanely complicated and eye wateringly expensive architectures for what are simple web apps on the big clouds. Often the team don't really understand the intricacies of all these services and how they interrelate which causes often cause a lot of stability problems.

Even if they don't, there is usually a huge tax on developer productivity with these architectures, which often outweighs any improved stability over the long term.

Let's keep in mind that modern hardware and software is very stable, generally.

Re: Hetzner Introduces ARM64 Cloud Servers

#284

Post by justsomehnguy »

nine_k wrote: post #273
justsomehnguy wrote: post #252

Earlier quoted context omitted.

Counterpoint: for some years we had clients who happily lived on cheap as fuck plans but requested more at the anticipated calendar events (NY and the like). Most of the time we just removed GHz limits, sometimes adding vCPUs in advance, they fared well. NB: IaaS provider with a guarenteered resources

Yes, if your elasticity requirements are not within hours but within weeks or months, adding and removing capacity can be done by traditional means, even by renting physical servers and putting them into your on-premises racks.

You missed the point - for some the 'elasticity' is just being able to consume 12-20GHz instead of ~1-2GHz 350 days of the year. Sure, you can deploy some advanced hyper-scaling arch to do so... or just move to a fatter instance.

Re: Hetzner Introduces ARM64 Cloud Servers

#285

Post by ChuffedToBits »

FinnKuhn wrote: post #179
2000UltraDeluxe wrote: post #108

Earlier quoted context omitted.

You have an address on your passport? I can see how that could cause issues if the address isn't valid. If I'd have to determine if your order was fraudulent or not, I'd err on the side of caution rather than risk having you abuse my server infrastructure (in the case of Hetzner).

German Passports have the City/Region you life in on them. The ID cards have your full address. So if the address of your ID card doesn't match your passport something is definitely fishy as you would have to get them updated when you move.

German (and most other?) passports feature the name of the issuing authority and place of birth, no other location, and remain valid irregardless of your current registered address.

Side note: It seems like many African ID cards list the profession and I've had to explain to several suspicious border guards why my passport doesn't.

EDIT: I've always wondered about this and apparently the body that sets passport specifications is ICAO, the International Civil Aviation Organization, as mandated by the UN. Specifically through ICAO document 9303.[1]

[1]https://www.icao.int/publications/documents/9303_p4_cons_en....

Re: Hetzner Introduces ARM64 Cloud Servers

#286

Post by derfabianpeter »

pachico wrote: post #145
andix wrote: post #129

Earlier quoted context omitted.

Hetzner also has cloud servers and an API to provision them. There is even a terraform module for hetzner cloud. It works very well. But they don’t provide any managed services, like databases or blob storage.

Yup, reason why we stay in AWS

If that’s the only reason, we offer managed databases of many flavors on hetzner. https://www.ayedo.de

Re: Hetzner Introduces ARM64 Cloud Servers

#287

Post by sshine »

nine_k wrote: post #220
zackmorris wrote: post #209

I implemented a Virtual Private Cloud (VPC) on AWS with Elastic Container Service (ECS) using Terraform so that I could run Docker, and it ended up being about 2 orders of magnitude more expensive than Hetzner after all of the services were configured. For example, something as simple as AWS NAT Gateway costs $30 per month, and it can be challenging to get everything right if you don't use one. I haven't tried Heroku…

Cloud is about elasticity. At ${work} we spin up a ton of EC2 instances in anticipation of huge traffic spikes that happen predictably several times a day, in different regions, and spin them down afterwards. It's also about simplicity. Many services come pre-configured on.AWS, with clustering, failover, backups, etc already present. When you are small, you can sysadmin your server fine, and you don't yet need the cl…

I can’t help but think that a 10x user spike causing a 10x CPU spike is mostly caused by poorly performing web apps written in dynamic scripting languages and the ethos “Just add more servers, servers are cheaper than developers.”

Turns out, developers are cheaper than DevOps engineers and AWS budgets. So if a linear spike in servers necessitates a superlinear cost in operational complexity, maybe working highly performant web app backends is actually cheaper?

Re: Hetzner Introduces ARM64 Cloud Servers

#288

Post by jamesfinlayson »

martinald wrote: post #283
glutamate wrote: post #279

Earlier quoted context omitted.

A simple single node is often more reliable than a complex fault-tolerant setup. Fewer moving parts mean fewer broken pieces. Similar with security. A simple monolith is more secure than isolated front- and backend.

Agreed entirely for most. I have seen so many insanely complicated and eye wateringly expensive architectures for what are simple web apps on the big clouds. Often the team don't really understand the intricacies of all these services and how they interrelate which causes often cause a lot of stability problems. Even if they don't, there is usually a huge tax on developer productivity with these architectures, which…

Agreed - I worked for a company with 30 people dedicated entirely to managing the mess they'd created in AWS - and their AWS account manager somehow managed to keep convincing them to use more and newer AWS services to dig them in even further.

Re: Hetzner Introduces ARM64 Cloud Servers

#289

Post by cellularmitosis »

ojosilva wrote: post #277
cellularmitosis wrote: post #233

Earlier quoted context omitted.

> simpler circuitry > the number of transistors tend to be similar between the two platforms sorry, come again?

x86 has more pathways (interconnects), x86 implements a "mesh" design, which is a low-latency, high-power configuration of pathways. ARM uses a neat hierarchical design that favors low-power consumption. So, given a x86 and an ARM chip with the same number of transistors, the ARM one will still be a simpler, low-power chip with probably a higher number of cores.

ah thanks! I hadn't considered the core count difference.

Re: Hetzner Introduces ARM64 Cloud Servers

#290

Post by watermelon0 »

daneel_w wrote: post #124
lopkeny12ko wrote: post #59

I don't understand the hype around ARM64 in the server space. Is this just due to all the marketing from Apple around their proprietary M processors leaking into the enterprise market? The latest Xeon and EPYC offerings are very performance competitive, I doubt we need to overhaul and entire processor ISA paradigm for continued improvements.

With Arm Graviton on AWS, for our specific environments and workloads, we get better performance for less money. Same goes for the hosts, who pay a smaller upfront investment per core and less in electricity per core. Bigger bang for the buck when the glove fits.

Same here, generally 10-40% better performance, depending on workload, at 15%+ lower costs, on AWS.

We have seen performance improvements in PHP, Java, and various CPU-bound tasks such as video transcoding.

This goes for either EC2 m7g vs m6i and also for Fargate x64 (from my experience, perfomance seems comparable to m5 instances) vs ARM (Graviton2).

Post reply on HN