Hacker News new | ask | show | jobs
by marktangotango 3949 days ago
Great perspective, not many people today have experienced the shear magnitude of mainframe batch workloads. The other thing about timesharing on the mainframe was system stability. IBM went to great lengths via cics to totally lock down stateful, bidirectional interaction with the mainfrme. Why was this? I've always wondered, why not time sharing?
1 comments
MichaelGG 3949 days ago
>not many people today have experienced the shear magnitude of mainframe batch workloads

Can you elaborate? For instance, SWIFT does like 15M messages/day (according to Wikipedia). That's...really not that much in absolute terms for even a cheap server today.

link
tolle 3948 days ago
But theres a ton of bookkeeping and checks to be done for every transaction. Due to regulations and so on. But yes it's not an insane amount of transactions.
link
marktangotango 3947 days ago
Note I specified 'batch'. One mainframe app I'm familiar with handled > 50% of mutual fund transactions in the US. Including 401k, this could be > 100 million transactons per day when follow on broker/dealer cascade transactions are accounted for. Each transaction required non trivial amounts of processing, reporting, and logging.

Also note that a 'transaction' in this sense is not a http request response. The way it's used in mainframe systems it's a business transaction which can include 100's of smaller 'transactions'.

Banks and other financial systems can't perform daily reconciliation until markets close and stock and fund prices are known. Hence these systems store everything up to process in a nightly window.

link