Cloud at Cost - Did I make a mistake?

scottalanmiller

What, can they stick like 2000 linux boxes on a single piece of hardware?
Certainly. And that number just increases with time. Every generation you get much higher density than the generation before.

scottalanmiller

@Dashrender said:

Same goes for the Windows side, I guess the density that they can provide much be much higher than I realized.

Yup. NTG has actually looked at when it would make sense to go to hardware that does close to 1,000 PBXs on a single 2U chassis. Virtualization takes you much farther than people realize.

coliver

@Dashrender said:

What, can they stick like 2000 linux boxes on a single piece of hardware?

Same goes for the Windows side, I guess the density that they can provide much be much higher than I realized.

If they are selling their developer 1 option predominantly (which I'm not saying they are) they could have that kind of density.

scottalanmiller

@coliver said:

@Dashrender said:

What, can they stick like 2000 linux boxes on a single piece of hardware?

Same goes for the Windows side, I guess the density that they can provide much be much higher than I realized.

If they are selling their developer 1 option predominantly (which I'm not saying they are) they could have that kind of density.

Those you could get 500 on a Dell R720xd (one generation old) which is a pretty low end, SMB server, with zero shared memory. Share memory and 1,000 on old hardware is doable!!

On an Oracle 2U box you might get 4,000 or more. On a big enterprise server, look out!

I'm guessing this is banking on the fact that 90% or so (guessing) of users will use next to nothing on resources and very few will be pushing it to the max. The average website I'm guessing ones the density is up is pennies to run.

scottalanmiller

@thecreativeone91 said:

I'm guessing this is banking on the fact that 90% or so (guessing) of users will use next to nothing on resources and very few will be pushing it to the max. The average website I'm guessing ones the density is up is pennies to run.

That's partially true, but not actually required for the concept to work. The overcommit level just makes it that much easier to work.

AmanBhogal

Hey everybody!

Aman from cloudatcost here.

was just checking the threads before calling it a night and came across this thread and really wanted to comment before getting some sleep.

There are a few things in this thread which are actually really viable questions... How does cloudatcost plan to support their servers? How do we pay for our bandwitdh? who's paying for all of the power of the datacenter? Will they be around next year?

These are all awesome questions and I'm really glad that they are being asked. There is a rational fear of trying something new, something different, something you havent heard of... and that's us... cloudatcost. I for one am really big on transparency and I believe that all of your questions should be answered, and I will do my best to answer them all right now:

First of all, let's start with who we are, how we were founded, and where are we going.

Cloudatcost is a brand of Fibernetics, a voice and data telecommunication provider in Canada. (Cambridge, Ontario... if you're local, come by and check us out for yourselves).
More information on Fibernetics is found here: www.fibernetics.ca

Fibernetics works in 4 major areas: Residential products (www.worldline.ca), business services (business.fibernetics.ca), wholesale, and finally the "Seedling Initiative" - a venture start up division of Fibernetics. http://www.fibernetics.ca/about-us/fibernetics-ventures/

as a side note... Fibernetics recently invested $2.1 million in a seedling called Plasticity Labs - a company thats focused around happiness in the workforce

So as you can see, our backbone is very robust. We have over 300,000 users alone on our residential products..

So the answer to how we can supply bandwitdh so cheap? Fiberneticss is a voice and data telecommunication company, these resources are shared across our brand, including cloudatcost.

How do we keep our operating costs so low?

The datacenter which you see in the youtube video I posted in another thread (https://www.youtube.com/watch?v=INVaNT1R_jE) is of our actual datacenter, not a staged location. The datacenter is powered using solar! we have solar panels over the entire roof - We capture so much sun that our hydro costs are virtually $0 - so lets pass the savings on.

Fibernetics is about being very disruptive in the tech industry, whether it be unlimited internet for residential users, robust business phone systems, or cloud servers - we do things differently, and we're awesome at it. Unfortunately, people think it's too good to be true.... what if they shut down shop and leave? - If I could do something to reassure you that we're not going anywhere, let me know what it is... Fibernetics is 10 years strong and growing at a pace like never before. cloudatcost is doing very very very well, our pricing model is different.... too good to be true... but it is true.

If you have any questions for me, let me know....

lance

I'm glad I ran into this thread, CloudatCost looks like some good stuff.

Well, after reading all of this, I am going to give it a go. I will have daily backups, just in case anyways

scottalanmiller

It's very low cost, I wouldn't buy fifty servers before you play with it. But grab one or two and see how it works for you, same as any other cloud or IaaS provider.

thanksajdotcom

My guess is that the company is banking on volume versus duration. They are counting on lots of the lower level plans, or even the higher-level ones. It's like how grocery stores make money. They lose money on basically anything on sale, but they figure that if the cost is right, people will just keep coming back.

For these guys, if a company's options are thousands to host their own internal network or use a VPN solution, like Pertino, as an example, (assuming a reasonably small business here) with their cloud servers they pay one-time for and then use for life, they can make up what they lose in recurring monthly costs in volume of servers people purchase for life.

I see this as a very viable marketing strategy, albeit a very different way of thinking compared to the norm. I like it personally. It's risky but I like it.

thanksajdotcom

Besides, look at it this way. Someone purchases lifetime servers at certain specs for Windows servers. 5 years from now, those specs are no longer adequate. Now they have to buy new lifetime licenses. The old servers get turned into Linux servers or dev servers. Everyone wins.

thanksajdotcom

Or they get decommed and Cloud@Cost wins. Either way, the company saves money, and C@C makes out well.

scottalanmiller

@thanksaj said:

Or they get decommed and Cloud@Cost wins. Either way, the company saves money, and C@C makes out well.

Yes, that is likely going to be a lot of cases.

scottalanmiller

@thanksaj said:

Besides, look at it this way. Someone purchases lifetime servers at certain specs for Windows servers. 5 years from now, those specs are no longer adequate. Now they have to buy new lifetime licenses. The old servers get turned into Linux servers or dev servers. Everyone wins.

Exactly. Buy a Dev1 today and that costs pennies to operate in the future and eventually becomes useless. Eventually no customers will keep those old systems up and running. "Forever" is only as long as practical.

QDesk

@Aaron-Studer said:

Theses link aren't helping:

http://www.theregister.co.uk/2014/01/15/hosting_outfit_nodeki_breaks_lifetime_promise/

http://www.webhostingtalk.com/showthread.php?t=278412

http://news.slashdot.org/story/12/08/17/1734250/joyent-drops-lifetime-account-holders

l think that it is important to not judge one vendor by the bad behavior of others. There will always be badly behaving vendors who pull a fast one, cheat you, steal your data or what have you. But you can't assume that everyone will do that or you would be unable to do business. If we looked at it that way, we would assume that because AMC went belly up that all car companies are going to go out of business and we would stop buying new cars because we are assuming that warranties are worthless because we are not evaluating the existing car companies based on their own merits but evaluating them based on the failure of an unrelated company.

I guess I should have trusted my gut...

Oh well, I got my money back.