ScalingBlockChains

ThoughtStorms Wiki

Context: BlockChain, IsMassNatural

VitalikButerin explains the issues and "Rollups" : https://vitalik.ca/general/2021/01/05/rollup.html (See also ZKRollups)

If I understand correctly, there are two "levels" of making blockchain "scale" (more transactions per second).

Level 1 : make on chain transactions faster

Level 2 : make some transactions off-chain, on side-chains etc. and then record some summarized version of them all on the main blockchain

Level comes in 3 versions :

  • StateChannels (AFAICT ... like prepaid "subscriptions" or "put it on the slate" eg. some money goes into a type of escrow, and participants make micropayments which are recorded, then finally, the total of microtransactions are added up and turned into a single transaction on the main blockchain) But only works for participants fully on the blockchain.
  • "Plasma" (this is complicated, but sounds more like a second accounting system which publishes automatic regular blockchain, which contain links into its own tree of transactions) So more like a separate currency kept synced to main chain. (Almost like a CurrencyGearBox)
  • Rollups (eg. ZKRollups)

Rollups move computation (and state storage) off-chain, but keep some data per transaction on-chain. To improve efficiency, they use a whole host of fancy compression tricks to replace data with computation wherever possible. The result is a system where scalability is still limited by the data bandwidth of the underlying blockchain, but at a very favorable ratio: whereas an Ethereum base-layer ERC20 token transfer costs ~45000 gas, an ERC20 token transfer in a rollup takes up 16 bytes of on-chain space and costs under 300 gas.

As blockchains face their scaling problem we move to ideas like https://steemit.com/informo/@johan-nygren/bitlattice-as-ethereum-2-0-and-how-sharding-required-a-new-state-topology which shard the blockchain into fragments and then rely on intermediaries to bring them back together.

Effectively Blockchains are moving to MiddleSpace

My Quora Answer (not sure how much of this still stands up, see elsewhere) : What do you view as the most significant risk for Bitcoins and Ethereum?

To me, it looks like the most significant “risk” is that the whole infrastructure is simply too heavy and unwieldy to fly.

The blockchain is too big for most people to keep a copy. The amount of data that has to be transferred every day becomes so big that it’s incredibly slow to keep your copy of the blockchain up-to-date. And if you can’t buy “unlimited” bandwidth at a fixed price, it might become so expensive that the cost of using these currencies is higher than the value you get from using them.

What happens then is that people start to use “bitcoin” or “ethereum” but from a “lite” wallet. Via a third party in giant data-centre, with mega-fast pipes and its own copy of the blockchain.

Once you do this, once you stop interacting with the blockchain directly, on your local machine, and start having to trust a third party to do your blockchain interaction for you, then why not just use a bank? You have all same problems of being dependent on your third party, of having no guarantee that they aren’t ripping you off, of the government coming and shutting them down, of thieves and hackers and excessive service charges.

Once you get to Ethereum, the problem is even worse, if I understand correctly, now you’ll need a lot of processing power as well as storage and bandwidth. Distributed storage systems like MaidSafe? Even more data flying around. If any of these things takes off and has to scale, they will inevitably lose their decentralized / flat / open nature and restructure themselves as hierarchical, with powerful local centres and gatekeepers.