Billmanager nulledcustomer_bills, insureds, and customer_currency) and provide access to aggregate info. We use cashflows and jurisdictional status-checks: when the bank needs to clear assets, we use statements to unlink assets. This lets them be reordered if the price changes due to events. We can also leverage the distributed systems that the banks use to track clearing, allowing transaction constraints to be lifted or removed. We also use zombies: when a bank needs cash, a zombie will be created and it can be used to jump on to the network.
We plan on use docking stations on the banks. Banks can use dock stations when needed. This will allow them to quickly move onto another system or perform other financial transactions while remaining on the network and avoid redundancy and the risk of missing an activity. This has the added benefit of going faster than a traditional system.
Banks are implementing a few different implementations, including one that mitigates the risk by using predictive modeling to determine the value of a transaction. In this implementation the bank includes a locking loop which involves a new set of servers, servers for each client, each server manage a book of transactions in a database, and another on each dock station. If this amount of work is done, the system can reduce the latency to 1-10 nanoseconds. The cost is around $2 million, and will last the first year of operation.
In addition to this I’ve been developing my own protocol for the banks, the Acharagi Bank, which is called APB. This protocol is based on Dockblock-based protocol.
One thing I’m not going to cover is how this implementations works: I will say it is difficult to deal with a large number of transaction fees and pass the cash. This is not an open source project. I would like to see this licensed, and how it can do the job better.
If you’re interested, please email me (also the bank’s help page can be found below).
Now, why do I care about blockchain? I hope that this blog post helps you to understand the technology an