This is a continuation of the earlier posts starting with this one: http://ravinote.blogspot.com/2020/09/best-practice-from-networking.html
Book-keeping is essential for both scheduler and allocator not only to keep track of grants but also for analysis and diagnostics.
A scheduler and allocator can each have their own manager that separates the concerns of management from their work
The more general purpose the scheduler and allocator become, the easier it is to use them in different components. Commodity implementations win hands down against specialized ones because they scale.
The requests for remote resources are expected to perform longer than local operations. If they incur timeouts, the quantum grants may need to stretch over.
Timeout must expand to include timeouts from nested operations.
Some event notification schemes are helpful to handle them at the appropriate scope.
A recovery state machine can help with global event handling for outages and recovery.
The number of steps taken to recover from outages can be reduced by dropping scoped containers in favor of standby
No comments:
Post a Comment