Is it plausible for some kind of anonymous organization having huge computational resources to attack on block chain by re-building entire block chain within the time frame of next block to be published and gain control over the currency.


I don't think the premise of the attack (rebuilding the entire blockchain within the time of one block) is necessary. If someone wanted to rebuild the entire blockchain, they could take, say, a year to do it, which would considerably reduce the computational requirements. Of course more blocks would be created during that time, so they would need to catch up at the end, but that would be much easier than re-creating the whole blockchain in 10 minutes.

The other thing they could do is start in the middle of the blockchain (unless you have a good reason for redoing the entire thing), which would also reduce the computational requirements. Or is the idea of rebuilding the entire chain that they would control all the bitcoins?

I think it would be reasonable for a three letter agency to build enough ASICs to have well over 51% and take over mining. (They could have enough capacity right now hidden in pools, waiting until they need to take control and nobody would know. And the hardware would pay for itself.) But rebuilding the entire blockchain would require an implausible amount of hardware for little benefit.

| improve this answer | |

No. Standard computer chips are magnitudes slower at creating valid blocks than ASIC miners. So, to rebuild 440,000+ blocks in the time-frame of one block, the attackers would need access to ASIC miners with the hashpower of ~90,000 times the current network's hashpower (because the earlier blocks were much lower difficulty, see comment of Nate below). Since that would be immensely expensive and production of such hardware would serve no other purpose than mining/attacking Bitcoin rebuilding the whole chain between two blocks is not a lucid attack.

| improve this answer | |
  • 1
    Note that many of those blocks were at much lower difficulty than the current level. My calculation is that you'd only need about 90,000 times the current hashpower. Still a lot, of course. – Nate Eldredge Dec 16 '16 at 2:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.