Nina is built on Solana which is a proof-of-stake blockchain. Proof-of-stake blockchains run on standard servers and do not need to run resource + power-intensive operations like proof-of-work blockchains. In fact, it has been posited that the cost of a transaction on Solana uses less energy than a single Google search.
Solana has recently published their 2023 energy use report, showing that the chain is carbon neutral.
Releases (but not bonus material) are stored publicly on Arweave. Arweave is a blockchain-based platform that focuses on permanent data storage. Its energy usage is primarily associated with the operation of its blockchain network, which employs a unique consensus mechanism called Proof-of-Access (PoA) instead of the more commonly known Proof-of-Work (PoW) or Proof-of-Stake (PoS) mechanisms.
We reached out to the Arweave team to get a better understanding of their energy consumption. Their response was:
Arweave's environmental impact is extremely low in terms of energy usage. Here's a breakdown of the main factors contributing to its energy consumption:
- The network uses at least 2 maximum capacity CPU threads for SHA-256 calculations, which are used for unlocking 'challenge entropy'. These threads are rarely more than a few at any given time.
- RandomX Hashing: Each packed partition requires 1 RandomX hash per second. Currently, there are around 32,000 packed partitions, requiring approximately 10 fully active CPUs globally. This level of CPU usage is so low that it can be easily handled by a Raspberry Pi.
- One-time Packing Process: When a miner initializes, there is a one-time packing process to secure the data. The energy used in this process is roughly equivalent to the energy needed to keep the data online for about 2.5 weeks.
- Drive Online Energy: There is also energy needed to keep the drives online, but this is relatively minor compared to CPU-related energy usage.
Without the packing process, the network would primarily rely on around 10 fully utilized CPUs and a small number of active threads, along with the energy needed for the drives. Considering the network's capacity and the efficiency of its operations, the energy footprint is significantly low.
More details can be found in this paper.