Mining will become controlled by botnets

[shads]

It's inevitable.  The widely held assumption is that the mining rate will stabilize around about the point where profits = energy costs + a nominal amount for new hardware investment.  But that assumes a level playing field.  Botnet operators will eventually build functionality into their user agents to tap into GPU's if they're available, when they do they'll have the power of a botnet behind them without power costs.  Botnets just became that much more profitable.

What are the implications of the block chain being under the effective control of botnet operators?  Well at least we can be confident they're unlikely to work together, competition in the black markets is probably more fierce than in legal markets.

[1713864179]

1713864179

[1713864179]

1713864179

[1713864179]

1713864179

[1713864179]

1713864179

[Xenland]

I thinkyou mean cpus not gpus i highly doubt operators will spend over $800 or more for shared hashing just to make a couple of buck  although pluasable.

[shads]

I thinkyou mean cpus not gpus i highly doubt operators will spend over $800 or more for shared hashing just to make a couple of buck  although pluasable.

No I mean GPUs.  If you've can run arbitrary code on slave PC its pretty trivial to detect the presence of a suitable GPU and send it the appropriate code to run.  Even to install any necessary drivers.  Why would the botnet operator be spending money? 

[Xenland]

I didnt mean to imply coding would be difficult.
But ive thinking about it a littlebit more and maybot net operators will hook up $800 gpus for faster resources and you purchase share of the gpu power. Personally i dont see the poit youd end up having to pay relativly the same amount becuase of how much the card costs. Id personally would rath spend $100 a month on saving up on gpus then spending it on a "share" of gpu resources. Own a video card can payit self off after a short amount of time and if mining gets unprofitable you could always sell the card and get some of your investment back

[shads]

You mean that botnets might rent out their processing capacity?  If so then I guess it's possible but it doesn't change the point, they will control large portions of the total mining pool at price point that makes it unprofitable for miners that have to pay their own energy costs.

[aceman1011]

This is old news. Many of my colleagues on HF are getting over 17 GH/s

[TheVirus]

Mining won't be controlled by botnets as GPUs will eventually be replaced by custom hardware (ASIC/FPGA). Difficulty will rise so high that having 20GH/s will be like having 20MH/s right now. Botnets will control a portion of mining for a short period of time, in relation to the BTC mining timeline. It's naive to think that GPUs will be the preferred method of mining 1-2 years from now.

[joker060]

If the difficulty where to jump that high, wouldn't most people stop mining?  Therefore dropping the difficulty back down?

[NetTecture]

If the difficulty where to jump that high, wouldn't most people stop mining?  Therefore dropping the difficulty back down?

No. You seem to have no idea how much that jump may be.

here something I was told:

ONE usb box starting availability soonish will take 50 watt and deliver around 8 gigahash.

3 6990 take about 1100 watt and deliver 2.4 gigahash.

Imagine

This will turn into a business side where you either invest into special gear, or get out.

[Auspician]

This pervasive myth needs to be shut down.  Mining will never be controlled by botnets for the following reasons:

A: Botnets can only easily gain control of computers run by those who are not technically savvy.  Non-tech-savvy people very RARELY run powerful GFX cards.  The Botnet owner would need thousands of computers in the Botnet to hope to achieve the power of a couple dozen powerful computers, or even a dozen duel-GPU systems.

B: Accessing the GFX card is one of the most difficult operations to do remotely, meaning this would only be achieved by a small minority of genius programmers capable of writing and constantly rewriting code to keep up with the nuances of security that gets in the way of accessing the GPU.  In almost all cases, even these Botnet owners would be lucky to access even half of the GPUs in their Botnet.

C: Mining requires very high levels of GFX performance.  The Botnet operator would need to either choose between having the GPUs in the Botnet run at full speed (causing both system slowness and crashes that would encourage the non-tech-savvy person to either buy a new computer or reformat the OS, likely increasing security as they do so, removing their computer from the Botnet).  OR, the Botnet operator would have to run the GFX cards at a slow enough pace not to interfere with performance on these lower-end machines, which means conservatively 25% utilization at maximum to avoid the owners resetting their systems to factory spec and wiping away the Botnet software.

Between these points, it quickly becomes clear that a Botnet consisting of 10,000 computers will not be nearly as effective as originally considered.  Point A takes the 10,000 and makes them as powerful as about 500 high-end dual-GPU systems, Point B drops this number to around 200, and point C drops the utilization to around 50.  Assuming 500 Mhash/s per system, that is a grand total of 25Ghash/s for the ABSOLUTE BEST and well run Botnet.  In all likelihood for the average Botnet owner (a few hundred systems) it simply wouldn't be worth the time or effort.  

Even if 100 'Absolute Best' Botnets came into existence (or a few less with more computers under their control), a change of 2500Ghash is only 20% of the existing network today.  Sure, these Botnet owners would make a profit, but if they all joined the network today (and none of them were previously in the network) we'd only see difficulty increase by 15-20%.  And its fairly safe to say that there are substantially less than 100 Botnets capable of running 10,000 computers with the absolute best software.

This delusion, however pervasive, should be outright ignored.  Mining will continue being hosted primarily by the network of GPU owners for the foreseeable future, so stop creating panic and paranoia.

[shads]

A: Botnets can only easily gain control of computers run by those who are not technically savvy.  Non-tech-savvy people very RARELY run powerful GFX cards.  The Botnet owner would need thousands of computers in the Botnet to hope to achieve the power of a couple dozen powerful computers, or even a dozen duel-GPU systems.

That's a pretty sweeping assumption... Are you saying all gamers are tech savvy?  I know plenty of die hard gamers that still think RAID-0 keeps their data safer because it's got RAID in the name.

B: Accessing the GFX card is one of the most difficult operations to do remotely, meaning this would only be achieved by a small minority of genius programmers capable of writing and constantly rewriting code to keep up with the nuances of security that gets in the way of accessing the GPU.  In almost all cases, even these Botnet owners would be lucky to access even half of the GPUs in their Botnet.

bollocks... I can download, configure and start a GPU miner from the command line.  From the command line I can check for signatures in the registry to detect the graphics card driver, I can download and install the appropriate OpenCL SDK.  None of this is hard for someone capable of writing the code for a botnet.  

C: Mining requires very high levels of GFX performance.  The Botnet operator would need to either choose between having the GPUs in the Botnet run at full speed (causing both system slowness and crashes that would encourage the non-tech-savvy person to either buy a new computer or reformat the OS, likely increasing security as they do so, removing their computer from the Botnet.  OR, the Botnet operator would have to run the GFX cards at a slow enough pace not to interfere with performance on these lower-end machines, which means conservatively 25% utilization at maximum to avoid the owners resetting their systems to factory spec and wiping away the Botnet software.

Detecting downtime for the machine is also not hard.  Any botnet operator worthy of the name would understand the risk of running the slave pc flat out 24/7.  Simply setting a low aggression with some kernels allows gaming without any noticeable impact yet when the PC is unused you get 95% of the hash rate you'd get if you had high aggression settings.  Your talking 10000 but the biggest known botnet today is 4 million.  Even it was only 25% utilization and they could get the GPUs working on only a fraction we're still talking massive processing power.

Between these points, it quickly becomes clear that a Botnet consisting of 10,000 computers will not be nearly as effective as originally considered.  Point A takes the 10,000 and makes them as powerful as about 500 high-end dual-GPU systems, Point B drops this number to around 200, and point C drops the utilization to around 50.  Assuming 500 Mhash/s per system, that is a grand total of 25Ghash/s for the ABSOLUTE BEST and well run Botnet.  In all likelihood for the average Botnet owner (a few hundred systems) it simply wouldn't be worth the time or effort.  

Even if 100 'Absolute Best' Botnets came into existence (or a few less with more computers under their control), a change of 2500Ghash is only 20% of the existing network today.  Sure, these Botnet owners would make a profit, but if they all joined the network today (and none of them were previously in the network) we'd only see difficulty increase by 15-20%.  And its fairly safe to say that there are substantially less than 100 Botnets capable of running 10,000 computers with the absolute best software.

This delusion, however pervasive, should be outright ignored.  Mining will continue being hosted primarily by the network of GPU owners for the foreseeable future, so stop creating panic and paranoia.

You can call it a delusion but you've made some very assertive calculations based on completely false assumptions.  

[Auspician]

If you have the expertise, go ahead and put together your own botnet and we'll see how well you do.

[shads]

If you have the expertise, go ahead and put together your own botnet and we'll see how well you do.

I have no interest in building a botnet.  I have more productive ways to use my time.  I just noticed you are in fact a pool operator.  I might remind you as I'm soon to be in the same boat myself that pool operators have no vested interest in the total number of miners, pool income is purely a function of market share with a small adjustment for the current rate of growth.  So I have no motive for scaring off miners or encouraging them.  I am simply interested in exploring possible future scenarios.  If you wish to be a tool and accuse me of spreading panic and paranoia so be it but at least try to understand the issue.

[Freakin]

There are not enough botnets to make a difference.  Someone making a massive botnet will get better profits from renting it out or using it for spamming.

This seriously is not a problem.

[bcpokey]

http://forum.bitcoin.org/index.php?topic=23243.0

This has all been discussed before.

[Serge]

deepbit can already detect botnets

[bcpokey]

deepbit can already detect botnets

Well first off, a pool has no vested interest in botnets. More hashing power to them means more money to them. Tycho has more or less been a standup guy in the bitcoin community, which is cool, but if botnets become more profitable I wouldn't be surprised or vexxed if he suddenly "stopped finding" botnets.

Secondly a botnet isn't exactly hard to detect. When one account suddenly has 5000-500000 IP addresses popping data in for it,  you'd have to be blind to not 'detect' it.

[Serge]

deepbit can already detect botnets

Well first off, a pool has no vested interest in botnets. More hashing power to them means more money to them. Tycho has more or less been a standup guy in the bitcoin community, which is cool, but if botnets become more profitable I wouldn't be surprised or vexxed if he suddenly "stopped finding" botnets.

Secondly a botnet isn't exactly hard to detect. When one account suddenly has 5000-500000 IP addresses popping data in for it,  you'd have to be blind to not 'detect' it.

I don't know how can anyone come to such conclusions. Hacking or seeding malware and taking control  of other people's computers without their consent is straight up illegal and criminal in most countries if not all. You think pool owner knowingly would want to be associated with such activity and profit off of it? [insert facepalm]

[xlcus]

Secondly a botnet isn't exactly hard to detect. When one account suddenly has 5000-500000 IP addresses popping data in for it,  you'd have to be blind to not 'detect' it.

If you were a smart botnet operator, could you not have all of your controlled machines mining for a bitcoind running on one of the machines?

[Auspician]

If you were running a Botnet you'd very likely want to run your own pool or simply have your bots run solo-mining, as any sane pool operator is going to take actions against you after realizing how many IPs are sending data for you. 

That said, the above objections stand, and in this individuals opinion it is seriously unlikely that botnets will ever account for more than 20% of the total mining going on. 


Just as a personal comment to shads, I don't mean to direct my words towards you specifically, but rather towards what I see as persistent propaganda attempting to scare away new miners.  I certainly encourage anyone concerned about this issue to do their research: I merely attempted to provide my interpretation of the botnet issue as it were.  Certainly, my numbers are just examples based upon my own estimation, and I would never claim any different.