Live data from Hacker News

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

cerebralab.com

131–140 of 338 posts

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

#131
post #109

Earlier quoted context omitted.

StackOverflow does predate the AWS world takeover.

How is this relevant? Amazon.com also predates AWS.

The implication is that they had already solved these problems and redoing the stack has its own cost to consider.

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

#132
post #12

One big thing not mentioned here: the massive collection of managed services AWS gives you at no cost. The biggest example is probably IAM, which makes it easy to control which servers can take what actions on which other services. And can also be integrated directly into external tools like Vault. Want to use service discovery? No need to set up Consul/Istio/etc., just use the managed service. Same with load balance…

> At no cost

That's so ridiculous I'm not sure how to reply or even interpret your comment.

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

#133
post #122

Earlier quoted context omitted.

> 20% minimum and typically much higher if you know how to negotiate when purchasing your on-prem gear. A 20% premium on a $10k/mo server bill is an incredibly small premium to pay versus a single FTE.

I have yet to move a single customer to Amazon that fired the FTE who was managing on-prem infrastructure and chalked it up to a cost-savings move. . The idea of cutting headcount is a fallacy. In small shops the guy that was managing on-prem infrastructure is now doing that (because it turns out you still need switches and phones and routers and firewalls even if you host things in AWS) as well as managing the AWS i…

We're a small company who avoided the hire. And we don't bother with a firewall at our office -- the only things on the office network are laptops, phones, printers, and sonos.

Basically, if you model senior eng time as $5k/week -- not even counting the opportunity cost -- you'll understand why AWS.

While I definitely think AWS is very expensive for the servers, they are not overall expensive. Again, setting up CloudWatch in an afternoon or dumping tons of heterogenous logs into S3 and querying them with Athena or spinning up full environment stacks (front end pool, back end pool, load balancers, databases, vpcs, vpns, etc) is an enormous cost savings vs finding, installing, configuring, and maintaining batches of software to do the same.

edit: pay an eng $180k. Fully loaded, you're looking at $220k plus. Add in either recruiting time or a recruiting fee that probably runs us $2k/mo amortized over the first year on the low end. Add in the fact that the first 6 weeks weren't super productive, and also absorbed tons of time from other SE to get this person up to speed. Add in PM time bringing this person up to speed on our product.

Yup, $5k/week.

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

#134
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…

Startups don't win because they save 50% on hosting costs. They win because they move fast on product development. AWS make the latter much easier due to all of the additional tooling they provide.

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

#135
I went into OVH and AWS rabbit hole. I think person needs to understand the concept of alternative cost when approaching such problem. Sure, you will save money on the server itself, but you don't get all other things that AWS is adding and you would have to one way or another do them by yourself. Even if your time is extremely cheap, I don't think one is able to beat their price as much to justify doing all this work. They've been at it for years and their solution is battle tested and you may be adding issues you won't be aware of that will bite you once you get growing.

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

#136
post #134
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…

Startups don't win because they save 50% on hosting costs. They win because they move fast on product development. AWS make the latter much easier due to all of the additional tooling they provide.

If the constituent employees are trying to win in the classical sense, and not the pump-my-resume-and-bail sense.

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

#137
post #63
post #13

With the price of pro fiber (redundant with SLA) I recently moved some apps back to our own servers in house. This did cut the price down dramatically. I would not recommend it for super critical apps (except if you have your own state of the art data center, but I am not speaking about that), but having 5-10 servers in a secure cabinet will give you infinite flexibility for little cost. We pay around 5000$/y to keep…

I like the idea, but remember that the SLA with your fiber provider doesn't mean your fiber magically will come back within a few hours when someone in the street cuts through the cable by accident. Your clients will still experience downtime unless you have two fiber lines going in different directions - completely redundant. If you don't have that, I would recommend some kind of a licensed microwave link on your ro…

We have a fallback coax connection going another way, it's only 500Mb/100Mb/s but it does the job if the fiber is cut. We are also working on having a 5G fallback just in case, as 5G bandwidth can be quite high on short range.

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

#138
post #111
post #15

We tried running things on AWS and digital ocean. Yes, AWS offers a lot of extra value with all the other services. But at the same time, we had a relatively simple application. We ended up managing our own sets of dedicated servers for a fraction of the cost. Peformance is way higher. In the end, the cost difference was just too big to justify it when we were trying to be scrappy. We also realized that all the extra…

So for you DO was cheaper, it's funny that I'm actually looking to move from DO to OVH, as DO pricing is around 50% more expensive than OVH. My main concern right now is the worse UI and the lack of more pre-built images that allow you to spin-up VPSs quicker.

Nope, it's OVH for us as well as this point.

We have about 8 or 9 dedicated servers at OVH right now. Things we've noticed:

- Yes, UI sucks

- No autobilling

- Reboot times are slow (as the actual machines are reboooted, not a VPS)

- You need to upgrade to higher bandwidths for great speeds

- Some routing problems, very rarely

- You really need to know what your server needs are 6-12 months in advance

- Some product offerings way too expensive (eg. Loadbalancer)

Great things:

- Price. Net-net, still cheaper to buy 2-3 dedicated servers than a single beefy VPS

- A lot of sales, like every month

- You get the full resources of the whole server

- Reasonable support for network or hardware related issues

- Unlimited traffic (at least you don't need to worry about normal usage overages)

- DDoS protection

Other thoughts:

In terms of performance, dedi's work great. I've found their OVH Public Cloud instances (not the cheap one, but the more expensive ones) have not as great performance. Might as well buy a dedicated. The only pro is that you only pay for the time you use.

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

#139

> 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.

Off-topic: Bandwidth is still pretty costly at Digital Ocean (DO). You can rent an entire droplet along with the additional bandwidth that comes with said droplet for 50% less than just buying the extra bandwidth (and even with that 50% discount it's still prohibitive for bandwidth-intensive apps).

More off-topic: I still generally like and use DO, but I dislike the policy of billing for one thing (e.g. 1vCPU) and adding hidden, fuzzy terms and conditions like "don't consume excessive CPU cycles." IMO the marketing ought to make it crystal clear that you're allowed to _burst_ to 100% use and also put in a well-defined threshold on any other resource limits like total CPU cycles so that people can plan accordingly and not be hit with surprise outages. It's similar to the complaint people have with Comcast selling XX00 Mbps packages and then tucking a miniscule bandwidth cap in the terms and conditions. That's potentially a fine policy, but it's extremely misleading to sell one thing and use fine print to shape it into a completely different offering.

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

#140
post #138
post #111

Earlier quoted context omitted.

So for you DO was cheaper, it's funny that I'm actually looking to move from DO to OVH, as DO pricing is around 50% more expensive than OVH. My main concern right now is the worse UI and the lack of more pre-built images that allow you to spin-up VPSs quicker.

Nope, it's OVH for us as well as this point. We have about 8 or 9 dedicated servers at OVH right now. Things we've noticed: - Yes, UI sucks - No autobilling - Reboot times are slow (as the actual machines are reboooted, not a VPS) - You need to upgrade to higher bandwidths for great speeds - Some routing problems, very rarely - You really need to know what your server needs are 6-12 months in advance - Some product o…

Thanks a lot for the information!

I was mostly looking to get some VPSs from them, the $10.58 2vCPU, 4GB RAM ones to replace the current DO $15/mo 2vCPU 2GB RAM I'm using, as I would get better specs for a cheaper price.

It's a bit hard to tell without actual benchmarks if the performance will be better or worse than DO.

Post reply on HN