Lennie Dymoke-Bradshaw wrote: >Looking through the various answers here, I note that someone has >mentioned the need for some kind of security, so that others cannot >trigger actions they are not supposed to. >I think those security issues conflict with your requirement to run >synchronisation without any authorisation.
I don't think that's a given. As long as the participants in this synchronization network can establish and maintain trust with one another, that should be fine even if none of them are particularly trusted by anything else outside their circle. Metaphorically it's like a secret society. DLTs (Blockchain networks) generally operate this way, for example. (I'm not necessarily suggesting a DLT-based approach in this case.) Have we learned yet whether or not it's acceptable if this synchronization network provides different answers if you ask different nodes the same question at exactly the same time? For example, if you have a 3 node network then hypothetically you could ask a question and if a majority of the nodes (2 out of 3) agree on a common answer, that's a valid answer. As another example, if the "truth" is 5 minutes out of date, that's OK, and a node will take some action if it isn't able to check in with its peers and maintain truth-consensus within 5 minute intervals. ....There are a LOT of possible options. - - - - - - - - - - Timothy Sipples I.T. Architect Executive Digital Asset & Other Industry Solutions IBM Z & LinuxONE - - - - - - - - - - E-Mail: [email protected] ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
