In a recent post on Reddit, a user discovered some anomalies in the hashpower of the DigiByte (DGB) blockchain, hypothesizing a 51% attack in progress.
Before going into detail, some aspects need to be clarified. This blockchain uses several algorithms, 5 to be precise:
- SHA256,
- Scrypt,
- Odocrypt,
- Skein,
- Qubit.
Some years ago a system called MultiShield was introduced, which allows calibrating the power, at least 20%, between the various algorithms as soon as a peak of hashpower occurs.
Looking at both the types of algorithms and the hardware with which it is possible to mine the crypto, both Scrypt and Qubit have values much higher than expected, and this suggests that someone has better hardware for that type of algorithms.
Looking closely at the blockchain it has been discovered that 2 addresses have mined hundreds of thousands of DGB in the last period, effectively monopolizing those two algorithms. Those addresses would have been able to achieve about 40% of the power of the network.
At this point, it would be enough to rent the remaining portion to allow a 51% attack on this blockchain.
The blockchain has become safer on the one hand by adding more hash power, but on the other, it is partially centralized by those who have managed to reach 40% of all its power without resorting to a pool.
One of the problems that troubles every blockchain is the issue of decentralization of consensus where, regardless of it, there can be either centralization of computing power in the hands of a few, using pools, or centralization in the same location. What changes is how attackable a blockchain is in the end and at what cost. The costs for an attack, in fact, can vary a lot, unless there is a coordination of the various miners.
