3 ms·
Sophisticated highly concurrent systems mitigate many of these problems by taking control of the scheduling order to dynamically optimize behavior. This elimina
by jandrewrogers 12y ago
Sophisticated highly concurrent systems mitigate many of these problems by taking control of the scheduling order to dynamically optimize behavior. This eliminates quite a bit of coordination and agreement that you see in a lot of concurrent software.
There is some interesting research on game theoretic adaptive schedulers for highly concurrent systems. Basically, if a scheduler in one process can model the behavior of schedulers in all other processes it interacts with, it can dynamically choose an operation with very low or zero probability of conflicting with the decisions made by all the other schedulers. In principle, this leads to all the schedulers getting maximum work done with minimum coordination or conflict.
Some algorithms I built for massively parallel systems work this way. At some level of concurrency, agreement and coordination becomes impractical so you have to come up techniques where the individual processes make choices that ensure they do not step on each other by making assumptions about the processes they interact with.