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

Is a billion dollars worth of server lying on the ground?

cerebralab.com

Post reply on HN
Print view View original

51–60 of 338 posts

Re: Is a billion dollars worth of server lying on the ground?

#51

Post by tluyben2 »

jdub wrote: post #9

Are there tens of regions around the world? Are there multiple AZs within a region that are single-digit milliseconds apart? Can I autoscale? Is there a managed load balancer? Do I have to manage my own database instances and availability? Do I have access to hundreds of other services for things that I don't want to spend time managing myself? Aren't we years beyond the 1:1 cost comparison for a single server argume…

No, almost no-one needs any of that. The depressing amount of money that goes to AWS for little reason but 'because it's the safe choice and everyone does it' is pretty amazing. In tech/programming we often say 'use the right tool for the job' and yet everyone reaches for the hammer that is AWS when hosting is mentioned.

Re: Is a billion dollars worth of server lying on the ground?

#52

Post by WrtCdEvrydy »

Cthulhu_ wrote: post #43
oriesdan wrote: post #7

I think the main reason they can afford pricing their services that high is because of peer pressure - probably itself the result of clever marketing, or that would be a really happy coincidence. I've worked in many startups now, several of them where I was the first (and for a while, only) developer and had to decide on the infrastructure. Each time I was going with OVH, and each time the CEO was trying to push for…

I've worked as a contractor for a CEO for two companies, in both he pushed for a full migration to AWS. Would not be surprised if he got a kickback from AWS. Amazon is pushing AWS pretty hard in the C-level, I don't know if you've ever followed one of their certifications or landing pages, but they do their marketing really well. Anyway, I do think a platform like App Engine / Beanstalk and other quick / easy / no se…

AWS allows you to shift your costs from CapEx to OpEx. Companies with low CapEx are valued higher since "theoretically" you could remove that bill by moving to another provider. Financial Engineering is just another part of software engineering and the cloud enables it.

Re: Is a billion dollars worth of server lying on the ground?

#53

Post by nikanj »

jdub wrote: post #9

Are there tens of regions around the world? Are there multiple AZs within a region that are single-digit milliseconds apart? Can I autoscale? Is there a managed load balancer? Do I have to manage my own database instances and availability? Do I have access to hundreds of other services for things that I don't want to spend time managing myself? Aren't we years beyond the 1:1 cost comparison for a single server argume…

YAGNI applies extremely heavily to most of those concerns. Pre-traction startups are wasting money they don't have, just to future-proof their application.

Re: Is a billion dollars worth of server lying on the ground?

#54

Post by api »

Bare metal destroys AWS and other big managed clouds and even smaller cloud companies like Digital Ocean if all you need is compute, storage, and bandwidth and want to manage it yourself.

Bandwidth is even more extreme than CPU/RAM. There are high bandwidth bare metal vendors that basically sell like rack vendors by capacity, not quantity, and offer bandwidth at stupidly cheap rates compared to any cloud. ZeroTier uses these for root server clusters and get tens of gigabits for The problem is that you have to manage it all yourself, which can be challenging and distracting, and these are no frills services with no add-ons like managed or specialized databases or S3.

That being said the cost savings are so extreme that you should consider it. It's ultimately a spreadsheet decision. Compare cost of hosting vs cost of DIY hosting plus labor for your specific application. Some apps benefit more from managed cloud than others.

There's also the multi-cloud option. You could host things that benefit from AWS or GCP there, and host the bandwidth or CPU-intensive stuff on bare metal.

Re: Is a billion dollars worth of server lying on the ground?

#55

Post by michaelbuckbee »

ehnto wrote: post #44
KingOfCoders wrote: post #18

Earlier quoted context omitted.

Many people use AWS because everyone uses AWS. Many of my clients have no need for AWS but still use it, at least until the VC money runs out. Usually then I have to go first, then servers are moving to somewhere cheaper when a new CFO comes in.

I work in an ecommerce agency environment and it can be quite frustrating. Every client, no matter how small, has been sold the spiel and had some guy come through and build a boutique cloud infrastructure, CI pipeline and development VM. As if traffic will increase 1000% by next week with no prior warning. It just doesn't happen like that. I spend half my time debugging Hyperscale Bullshit™ that the client never nee…

I did work for a YC startup that replaced my ~$100/mo Heroku app with a $5,000/mo AWS stack that nobody in the place knew how to manage.

Their site was so ridiculously low traffic that I did some napkin math and figured out they were paying around $0.10 per web request.

Re: Is a billion dollars worth of server lying on the ground?

#56

Post by darkwater »

viraptor wrote: post #10

He kind of touches on this in point IV, but never mentioned specifically: the cost comparison is not AWS vs ovh. It's fragment-of-EC2 vs ovh. If you need a predefined small number of VMs and no other functionality, it would be silly to go with AWS. But on the other hand, if you want a set of servers of a given class spawning on demand, with traffic coming in via load balancers, with integrated certificate and DNS man…

Absolutely. But they also charge you independently for (almost) all those services. So, you are somehow still paying for a sort of "comfy lock-in" (no native speaker here so try to follow me even if words are not the exact ones) because you cannot access those services from another pure CPU resource on another (cheaper) provider.

Basically, AWS is squeezing as many dollars as possible from your pockets (good for them).

Re: Is a billion dollars worth of server lying on the ground?

#57

Post by ckdarby »

oriesdan wrote: post #7

I think the main reason they can afford pricing their services that high is because of peer pressure - probably itself the result of clever marketing, or that would be a really happy coincidence. I've worked in many startups now, several of them where I was the first (and for a while, only) developer and had to decide on the infrastructure. Each time I was going with OVH, and each time the CEO was trying to push for…

Being on AWS, Azure or GPC isn't about the costs being pressured from another department at all.

The cost reflects the premium for integration into the other services.

It is after first employees bounce or the company outgrows that hustle hack everything together mentally that consultants like myself get hired and want to sob that some employee decided to board to OVH as a small startup especially if it is growing.

Every time I've dealt with this it is a disaster or things hanging together by a thread. Hanging together by a thread have sort of gotten better with kubernetes mostly keeping everything running just by turning anything back on when it dies or crashes. The thing is that only until last year did OVH offer managed k8s, those poor startups that will suffer from these choices.

OVH has it's place to be considered:

+40 engineering companies

High out going bandwidth (CDN, video streaming, etc.)

IO Latency requirements

Large enough scale of anything where the cost of AWS egress bandwidth is too costly

Re: Is a billion dollars worth of server lying on the ground?

#58

Post by jonstewart »

As a developer at a big company, if I try to buy a server, then I have to deal with my IT department. I don’t get to buy what I want, have to deal with particular overpriced vendors, and the process slows to a crawl.

If I want to use AWS, however, I get instant gratification. And of course I can experiment with different models and then refine into cheaper service mixes as scale increases. As base load emerges, I can use reserved instance pricing and shift that onto our capital budget. And of course I can control everything with code using AWS CDK.

At a smaller company where I was able to control the IT process more, it might be worthwhile. But the speed and optionality I get with AWS is vastly more cost effective than dealing with legacy enterprise IT processes.

Re: Is a billion dollars worth of server lying on the ground?

#59

Post by jrochkind1 »

michaelbuckbee wrote: post #55
ehnto wrote: post #44

Earlier quoted context omitted.

I work in an ecommerce agency environment and it can be quite frustrating. Every client, no matter how small, has been sold the spiel and had some guy come through and build a boutique cloud infrastructure, CI pipeline and development VM. As if traffic will increase 1000% by next week with no prior warning. It just doesn't happen like that. I spend half my time debugging Hyperscale Bullshit™ that the client never nee…

I did work for a YC startup that replaced my ~$100/mo Heroku app with a $5,000/mo AWS stack that nobody in the place knew how to manage. Their site was so ridiculously low traffic that I did some napkin math and figured out they were paying around $0.10 per web request.

Huh, making a manual AWS stack 50 times more expensive than a heroku setup seems like a real trick to me! I'm used to heroku being more expensive than running it yourself -- as I would expect, since you're paying them for a lot of management.

Re: Is a billion dollars worth of server lying on the ground?

#60

Post by andrenotgiant »

> On the flip side of the coin, there are server providers such as digital oceans, GC, and Azure that can be more expensive than AWS.

What gave the OP the impression that Digital Ocean is more expensive than AWS? AFAIK there is no configuration of Droplets vs EC2 where Digital Ocean is more expensive. In most cases, especially with any outbound bandwidth, it is cheaper by a large amount.

Post reply on HN