Scrypt Cloud Has Been LAUNCHED!!!!
-
[quote name=“ghitzafunny” post=“49317” timestamp=“1388703927”]
can I get only 1 megahash for a month , or 200 kilohash for 6 months ? I don’t have enough for 1 megahash for 6 months :| . Thanck you so much and sorry for spaming on the giveaway topic :D
[/quote]PM sent :)
-
Just wanted to say thanks for the big reply earlier :)
-
Ok so due to a lot of requests for shorter contracts I am now offering 3 month contracts.
1Mh/s - 0.55 BTC
2Mh/s - 1.1 BTCBut really this is the best I can do with manual processing for now…
-
a curious question about your pricing model. Will your prices fluctuate together with the currencies (btc, ftc whatever you) value, or will you always be flatrated at the prices mentioned for the contracts?
Normally, a cloud service offers less risk and less work to the customer buying the service, but here you basically put all the risk on the customers and none on yourself. If no one buys a contract, you still have a lot of power and the cards can ROI at their own pace. If a person buys a contract at the price of say 1BTC, they take a risk since the difficulty can rapidly skyrocket, thus leaving them with a service that costs 1BTC over 3 months, but only generate half of that back (since they buy a static ammount of MH). Meanwhile, you still get the income, either from contracts or by just mining with whatever power you have left over.
Wouldn’t it be more constructive (and make more sense) and win-win to take payment in percentage of the profit that the customer generates? :)
-
[quote name=“vizay” post=“49476” timestamp=“1388745758”]
Wouldn’t it be more constructive (and make more sense) and win-win to take payment in percentage of the profit that the customer generates? :)
[/quote]Nope. That puts the risk back on the host. It makes the most sense to maximize your profit potential by putting all the risk on the buyer. What you suggest is a win for the customer, and a loss for the host, and there’s no incentive to do it.
-
I totally get that, but I can’t for the love of god see why a customer would want to buy a service at risk when he can basically do the exact same thing, at a better profit, and without the risk? :)
Kind of drifting towards the cloud discussion again, I have a hard time seeing this being worth the time put in from FTClover. If you can’t over-allocate the resources the benefit just isn’t big enough imho.
-
[quote name=“vizay” post=“49552” timestamp=“1388776366”]
I totally get that, but I can’t for the love of god see why a customer would want to buy a service at risk when he can basically do the exact same thing, at a better profit, and without the risk? :)Kind of drifting towards the cloud discussion again, I have a hard time seeing this being worth the time put in from FTClover. If you can’t over-allocate the resources the benefit just isn’t big enough imho.
[/quote]Doubling down.
0. Calculate what percentage each machine can do and correlate how many days it would take to ROI for each fraction of a config.
1. Buy a massive amount of hardware. On credit.
2. Sell the contracts, as he’s doing here, for BTC.
3. Hold the BTC until XX% increase. Use this to buy near equivalent hardware. Allocate this hardware as “the house”, will not be contracted out.
4. “The house” mines directly into payment to go towards the credit. After attaining ROI on the first set, hardware capacity now doubled, all equipment paid for.
5. ? ? ?
6. Prof-OMG GUYS DON’T ACTUALLY DO THIS. -
[quote name=“HopeStillFlies” post=“49554” timestamp=“1388778122”]
[quote author=vizay link=topic=6481.msg49552#msg49552 date=1388776366]
I totally get that, but I can’t for the love of god see why a customer would want to buy a service at risk when he can basically do the exact same thing, at a better profit, and without the risk? :)Kind of drifting towards the cloud discussion again, I have a hard time seeing this being worth the time put in from FTClover. If you can’t over-allocate the resources the benefit just isn’t big enough imho.
[/quote]Doubling down.
0. Calculate what percentage each machine can do and correlate how many days it would take to ROI for each fraction of a config.
1. Buy a massive amount of hardware. On credit.
2. Sell the contracts, as he’s doing here, for BTC.
3. Hold the BTC until XX% increase. Use this to buy near equivalent hardware. Allocate this hardware as “the house”, will not be contracted out.
4. “The house” mines directly into payment to go towards the credit. After attaining ROI on the first set, hardware capacity now doubled, all equipment paid for.
5. ? ? ?
6. Prof-OMG GUYS DON’T ACTUALLY DO THIS.
[/quote]I will just say that this is a very profitable venture, not so much short term but more long term, initial sales exceed’s the initial cost’s by 1/3 so money invested is made back + 1/3 “profit/developement money” to put back into it so I can then offer a better service, and more mh/s, after a year or so I will then be CEO of a company worth millions! Also I am not selling all 31.5mh/s, 1.5mh/s or so is mine to cover electricity and maintenance.
Think of it this way and sorry for using facebook as an example but do you think mark zuckerbourg made millions the first month facebook was launched? nope it took months for popularity to gain and then it became profitable.
The real fun begins when scrypt Asic miner’s are launched and I will be able to offer much more mining power for a lot less, all expenses will decrease and also a lot less maintenance will be involved.
I have been talking with zerodrama about making the site more automated and including features such as user accounts and a mh/s market where users can trade/sell/buy mh/s. So once I put in all the hard work in the beginning it’s just smooth sailing from there on out :)
-
Would it also be possible to add an ability to select a pool?
-
Like the idea behind it !
Did even thought of doing this myself and offering hash-shares to friends and family (maybe rethink about it again…)Some questions though:
Why are you running Windows as an OS ?
Why not running your own P2Pool server(s) ? -
[quote name=“prensel” post=“49663” timestamp=“1388827932”]
Like the idea behind it !
Did even thought of doing this myself and offering hash-shares to friends and family (maybe rethink about it again…)Some questions though:
Why are you running Windows as an OS ?
Why not running your own P2Pool server(s) ?
[/quote]The only OS I am familiar with is Windows and also I can get it for free so why not :) and I wouldn’t know how to run my own pool :-/
And to keep things simple and have every rig doing the same thing near enough I want to mine with specific pools with one worker for each cryptocurrency then split the coins mined into shares according to mh/s :)
-
[quote name=“FTClover” post=“49697” timestamp=“1388837453”]
[quote author=prensel link=topic=6481.msg49663#msg49663 date=1388827932]
Like the idea behind it !
Did even thought of doing this myself and offering hash-shares to friends and family (maybe rethink about it again…)Some questions though:
Why are you running Windows as an OS ?
Why not running your own P2Pool server(s) ?
[/quote]The only OS I am familiar with is Windows and also I can get it for free so why not :) and I wouldn’t know how to run my own pool :-/
And to keep things simple and have every rig doing the same thing near enough I want to mine with specific pools with one worker for each cryptocurrency then split the coins mined into shares according to mh/s :)
[/quote]Oh man, i think your a bit over your head.
How do you know how much each miner contributes to the total share count? Mhash do not reflect Accepted shares at a pool, this might vary, how do you divide this.
What about the pool fee?, it seems it would be more simple imo. to set up linux, activate ssh and give the “renter” the ability to mine what he wanted and to where he wanted.
But then again you are up with troubles regarding users pushing it to high, crashing rig etc.So the best would be to just get an persons user account name and worker on a pool and then you direct the power there.
So when you say you are renting out 1Mhash you are renting out 1 and 1/3 of a 280X card?. would it just be much more simple to rent away x amount of cards with promise of around X khash performance pr card?
-
I am watching this with some interest.
-
I’m going to come forth as a pessimist here…again…
Just know that I’m doing that to be friendly, even though it might sound like I’m just a pessimistic douchebag. ;) I’ve seen friends in the crypto business (mainly around btc) that got in over their heads, and it turned nasty pretty quick (dealing with other peoples money is always risky, sooner or later you get tangled up with the wrong kind of people)aanyways…
Have you thought anything about network & IT security for this project? A single miner here and there never draws much attention, but when you come forth as a service provider with a whole farm like yours, you’re going to attract attention. How will you handle that kind of situation?
TL;DR
How are you going to prevent people from DDoSing and hacking the living sh*t out of your mining farm for their own purposes? -
You mention that you’ve got friends that might be able to help you locate the equipment? Why not setup 2 separate mining farms?
Sure you’ll have to do everything twice, but it seems to me that you’ll end up with 2 smaller (ultimately more manageable) mining farms, rather than one scarily big one.
In a residential context 40 GPU’s are a horrible supply/cooling headache.
Even 1 x 20 is frightening to me (let alone 2 x 20), but it somehow seems more manageable?
Also you don’t have all of your eggs in one basket.
Or is it the same difference? lol… I’m kinda tired. Overall want to simply offer some encouragement :P
[quote]So when you say you are renting out 1Mhash you are renting out 1 and 1/3 of a 280X card?. would it just be much more simple to rent away x amount of cards with promise of around X khash performance pr card?[/quote]
+1 In light of the fact it is not trivial for you to carve this stuff up, I think that makes a lot of sense.
The one thing I know for certain is that people in this community won’t let you fail, if at all possible This kind of thing sets FTC above the rest.
-
If he splits up his farm to a friends place he would have an extra fail point in some terms.
Scenario:
One of your rigs does a hard crash, your not able to ssh into it or use any other form of remote management.
You´d better hope that friend is home. Cause the only solution to this is a hard restart. So either you gotta have 24/7 access to that place, or you need to get yourself an PDU unit to control turning power on/off remotely over WAN. That again requires that the computer doesn’t boot into some failsafe stuff, i occasionally have to find my monitor and keyboard for my headless miners for manually getting the computer to load the operating system. You could automate this. but that doesn’t always work from my experience.There is like said above the case of DDOS issue, unfortunately theres not much to do about this, don’t show your ip address, or hide behind an relay address that you could change if it gets to hot out there. On a positive side, ddos attacks usually don’t last for a longer period of time.
-
@FTCClover - really a cool idea and I wish you success in setting it all up. Your page already looks promising - really well done +1 from me … let’s see how it will evolve … I’m just short before buying a contract as your pricing model is really fair.
Just on small info - I was not that convinced from using the D2 pool alone … had some downtimes that were not that nice. I made good experience in having GiveMeCoins as my primary pool and D2 as Fail-Over configured (or maybe even distributing your load across multiple pools… this way you balance pool downtimes).
Wish you good look in your current setting it up phase.
-
[quote name=“svennand” post=“49908” timestamp=“1388913234”]
If he splits up his farm to a friends place he would have an extra fail point in some terms.Scenario:
One of your rigs does a hard crash, your not able to ssh into it or use any other form of remote management.
You´d better hope that friend is home. Cause the only solution to this is a hard restart. So either you gotta have 24/7 access to that place, or you need to get yourself an PDU unit to control turning power on/off remotely over WAN. That again requires that the computer doesn’t boot into some failsafe stuff, i occasionally have to find my monitor and keyboard for my headless miners for manually getting the computer to load the operating system. You could automate this. but that doesn’t always work from my experience.There is like said above the case of DDOS issue, unfortunately theres not much to do about this, don’t show your ip address, or hide behind an relay address that you could change if it gets to hot out there. On a positive side, ddos attacks usually don’t last for a longer period of time.
[/quote]Splitting up the farm has it’s pros and cons for sure. You mentioned a big con, but a big pro is that he’s less sensitive to power outages, still being able to deliver partially to customers, which is a good thing.
The remote PDU part is pretty easy to build, all you need is a RasperyPi and some cheap electronicl components, together with some pretty basic knowledge of how to connect it all and script a solution for it. You piggyback the PI-solution on the powerButton for the rig. Once that’s done, create a script that pings the rig, when you loose 10 or more pings (or so) send a signal to restart the computer and pause the signal sending until a new ping is recieved (indicating the rig is up again).When it comes to security it’s a bit more tricky. In order to protect yourself from DDoS you need a secondary internet connection, a pretty descent firewall and a pretty descent router together with some knowledge on how to set up proper routing, rules and protection (forget home routers, you need something more powerfull like a routerboard 2011 or even more powerful).
What you do is keep a secondary internet connection like a backup, and when the primary connection get’s attacked you fail over to the secondary. This gives you time to get help on your primary connection from your ISP. It’s a cheap and easy to maintain solution to buy you time and avoid downtime with. However, if the attackers figure out where your secondary connection is you’re screwed anyway, but that should take some time at least :)
There are more rigid ways to solve the DDoS problem more permanently, but that costs more than it’s worth imho.The biggest threat you have to protect against is someone breaking into your farm remotely. If that’s done, one can easily steal all your power or use the machines for other purposes which is bad bad bad! First of all don’t run windows. Windows 7 & 8 is pretty secure, but history tells us that when a new exploit surfaces, it takes some time for microsoft to act on it and patch the vulnerability. Linux distributions like Debian, CentOS and openSUSE have a better track record here, way better!
This is also a good reason to have a powerfull firewall, you need to shut down EVERYTHING that doesn’t need to traverse the networks. On top of that you need some kind of monitoring that can send you alarms on everything from rigs going down to attacks on your network. -
I guess it depends on access, but presumably 24/7 access would be a requirement.
[quote name=“svennand” post=“49908” timestamp=“1388913234”]
If he splits up his farm to a friends place he would have an extra fail point in some terms.
[/quote]That is a major point against splitting the deployment between 2 locations.
I’m not experienced enough in the actual deployment of this stuff, but it just seemed to me that this option at least halves the technical challenge of powering and cooling 39 co-located GPUs.
If it causes more problems that it solves, then its not worth the bother.
-
Hi, it sounds like a good idea. About the security issue: wouln’t it help if you distribute the coins every day to your clients ? It should minimalize the safety risk if you reduce the amount on your account . Despite it will take more work for you it can help to get the confidence of customers.