I am curious, if I may ask, why is it so memory hungry? From my limited understanding, while downloading the blockchain it is also validating it. This has to be done one block at a time, because of the nature of the blockchain itself (I guess I can validate a random block, but at some point I will need a hash of the previous, which I must check is valid before I can say this one is valid). If each block weights 1 MB, then how are we needing > 8 GB?
The blocks directory contains all confirmed transactions. The chainstate directory contains all knowledge on all existing UTXOs ("funds"). With each block Bitcoin Core downloads, it verifies all transactions in that block, and needs to check if no UTXO that doesn't exist gets spent (as that would mean Bitcoin is created out of thin air), and it updates chainstate by removing the spent UTXOs and adding the newly created ones. So for each block, it has to check (up to) thousands of UTXOs, and chainstate tends to grow over time. So if it comes from RAM, it's quite fast. If it doesn't come from RAM, it has to read and write thousands of UTXOs from disk for each block it verifies.I'm not a blockchain expert, so this is my layman interpretation
