What wouldn’t it price to settle Korea Trade’s each day buying and selling worth on Ethereum? On the baseline, block-gas capability just isn’t the binding constraint. Settlement structure and new onchain state matter extra.
This text fashions a settlement workload on Ethereum calibrated to Korea Trade (KRX)’s reported common each day buying and selling worth, beneath present mainnet guidelines and beneath Glamsterdam’s proposed gasoline repricing. It’s a gas-accounting experiment, not a proposal to maneuver Korea’s settlement infrastructure onto Ethereum.
Capability just isn’t the issue on the baseline. At an assumed $10,000 common commerce, settling each commerce in its personal transaction would use 26.0% of at some point’s theoretical block gasoline at right this moment’s 60-million-gas restrict. That’s the full-day gasoline ceiling, not spare capability left after Ethereum’s present demand, and it reaches 100% if the common commerce falls to about $2,604.
Value construction is. If clearing first nets trades in order that one participant place per commerce is left to replace, netted settlement is 3.6× cheaper in gasoline than trade-by-trade settlement right this moment, however 2.6× beneath Glamsterdam. Glamsterdam reprices storage writes, and new state most of all. As soon as netting has eliminated a lot of the settlement work, the storage created for first-time holders turns into a bigger share of what stays.
1. Measurement the market
In an April 2025 article for the World Federation of Exchanges (WFE), KRX’s Chairman and CEO reported a median each day securities buying and selling worth of $21.34 billion (2024 knowledge) and an 89% netting ratio. That ratio measures how a lot gross worth clearing removes earlier than settlement, not what number of positions would stay onchain.
Dividing $21.34 billion by an assumed $10,000 common commerce offers 2,134,000 trades on a consultant day. Commerce rely is derived from worth, not from KRX’s individuals, devices or settlement cycles. The share of this workload moved onchain is swept from 0% to 100%.
Commerce measurement is the belief that strikes capability most, as a result of commerce rely scales inversely with it. Throughout averages of $2,000–$50,000, gross settlement with one transaction per commerce wants between 130.2% and 5.2% of theoretical each day block gasoline at full migration. It reaches right this moment’s ceiling at about $2,604 per commerce; batching 100 trades per transaction strikes that threshold to $1,598.
The workload is a median day, not transaction-level KRX data or a stress-day replay.
2. Change the settlement structure
Each architectures go away commerce matching offchain:
- Gross DvP (supply versus cost): each commerce settles atomically onchain, linking the safety and money legs. Trades might be submitted one per transaction or batched.
- Netted DvP: clearing offsets every participant’s buys towards their sells offchain; every participant then settles one internet place onchain per netting cycle.
Netting is best to see on 100 trades. Every has a purchaser and a vendor, so gross settlement updates 200 participant positions. If the identical brokers seem on many trades, netting leaves solely their internet adjustments to write:
I name this ratio P/T: internet participant positions up to date onchain per gross commerce. All three rows are assumed situations, not noticed KRX place counts, and KRX’s 89% worth ratio doesn’t determine P/T. The place KRX would sit is determined by the account hierarchy: clearing-member or custodian accounts consolidate many trades, whereas end-investor settlement preserves much more positions.
To cost every structure, I measured a minimal balance-settlement contract on a neighborhood node beneath present mainnet guidelines (Prague). Holders had been funded in earlier transactions; gasoline got here from transaction receipts and was checked towards opcode traces. The one-trade row is a full transaction, together with intrinsic gasoline and calldata. For batches, the fastened overhead per transaction is separated from the incremental price per unit, which covers that unit’s execution and calldata.
Netting benefit: whole gross gasoline ÷ whole netted gasoline, with no new holders. Glamsterdam columns are modeled, not executed.
The contract settles balances solely. Custody, identification and compliance checks, company actions, and restoration logic would every add storage reads and writes, so a manufacturing token would probably price extra per operation. The experiment additionally doesn’t evaluate Ethereum mainnet with L2s or permissioned settlement networks.
3. Capability on the baseline; state prices on the margin
Glamsterdam adjustments two issues. Its repricing makes the identical settlement workload heavier in gasoline; solely the next gasoline restrict turns that into extra headroom. And it fees much more for creating state than for updating it, which issues most to the structure that updates the least.
The situations use present guidelines at 60 million gasoline per block, and Glamsterdam repricing on the identical 60-million restrict and at 150 million, a devnet reference relatively than a scheduled mainnet restrict. Glamsterdam’s mainnet activation date is unconfirmed, and no Glamsterdam EVM is executed right here: the measured storage reads and writes are repriced beneath the proposed EIP-8037 and EIP-8038. EIP-8037 meters execution gasoline and state gasoline individually; a transaction pays for each, however every has its personal block restrict, so utilization follows whichever binds.
At full migration, one-transaction-per-trade gross settlement makes use of 26.0% of theoretical each day block gasoline right this moment. Glamsterdam’s repricing alone raises that to 37.3% on the identical 60-million restrict; at 150 million it falls to 14.9%. Batched gross settlement makes use of 16.0%, 30.2% and 12.1% in the identical three instances.
The repricing reaches each sorts of storage write:
- Updating an present stability. Beneath EIP-8038, the primary write to an present storage slot after a stability learn rises from 2,900 to 10,100 gasoline on this contract. A gross commerce between present holders rises from 52,722 to a modeled 75,522 gasoline.
- Creating a brand new holder. Beneath EIP-8037, a brand new storage slot prices 97,920 state gasoline plus 10,100 execution gasoline, towards 20,000 gasoline right this moment. The contract writes two slots per new holder, a stability and a first-holding flag, so every one provides 10,310 execution gasoline and 195,840 state gasoline beneath Glamsterdam, towards 37,310 gasoline beneath present guidelines.
On this mannequin, each architectures face the identical new-holder inhabitants. As a result of netting reduces the opposite settlement work, new state accounts for a bigger share of its remaining gasoline. With one place per commerce, the benefit falls from 3.6× to three.1× as new holders rise from 0% to 10% of trades right this moment, and from 2.6× to 1.9× beneath Glamsterdam. With extremely concentrated netting, it falls from 33.0× to 10.6× right this moment and from 23.8× to 4.0× beneath Glamsterdam.
All comparisons are in gasoline. Greenback charges depend upon gasoline costs, and settlement latency or failure danger is determined by a system design this experiment doesn’t mannequin.
Closing
Beneath these assumptions, block-gas capability doesn’t cease Ethereum from settling a workload of KRX’s each day buying and selling worth; the price of state shapes which settlement design is cheaper. Netting saves gasoline by lowering the positions that want updating. However when a first-time holder nonetheless wants a brand new onchain stability, that storage creation stays costly beneath Glamsterdam’s proposed pricing. The extra environment friendly the netting, the bigger the share of its remaining price that new holders account for. A Glamsterdam with the next gasoline restrict would subsequently add capability whereas narrowing netting’s relative gasoline benefit.
How far it narrows is determined by two numbers this experiment needed to assume: what number of positions netting actually leaves, and what number of new holders a tokenized market brings in. The capability consequence additionally is determined by the $10,000 common commerce and on batching, and the Glamsterdam figures on EIP-8037/8038 parameters that aren’t but closing. Participant-level netting knowledge and an noticed new-holder charge from a reside tokenized-securities venue are the following measurements price making.
Appendix
- Code and knowledge. Gist (gas-measurement harness, Glamsterdam repricing, workload development, and consequence tables).
- Fig. 2 particulars. Strong traces use two new slots per holder, as within the measured contract; dashed traces use one. Gross settlement is one commerce per transaction, and each architectures obtain the identical new holders. The left panel stops at 10% as a result of extra new holders than 0.11 internet positions per commerce is infeasible.
- Associated. What Adjustments When Tokenization Reaches Market Scale?
This put up was initially revealed on my private weblog: https://egpivo.dev/2026/10/11/what-glamsterdam-would-change-about-settling-korean-stocks-on-ethereum.
What Glamsterdam Would Change About Settling Korean Shares on Ethereum was initially revealed in The Capital on Medium, the place persons are persevering with the dialog by highlighting and responding to this story.
