With the author's solution, we'll have events with the same vector clock
In "Virtual Time and Global States of Distributed Sysmtes" page 126 says A process Pi ticks by incrementing its own component of its clock: Ci [i] := Ci[i]+1 As before, ticking is considered to occur before any other event action, and the timestamp C(e) of an event e is the clock value after ticking.
In "Scalability Approaches for Causal Multicast: A Survey" page 7 saysEach time a message m is multicast by pi , it is tagged with the local vector clock of its sender (i.e., VCm := VCpi ), once such vector clock has increased its local entry (i.e., VCpi [i] := VCpi [i] +1).
How to know if we can deliver a message?
(TODO: someones compare with <= and others say that is needed the +1 rule, see Scalability Approaches for Causal Multicast: A Survey)
When to increment the counter
With the author's solution, we'll have events with the same vector clock
In "Virtual Time and Global States of Distributed Sysmtes" page 126 says A process Pi ticks by incrementing its own component of its clock: Ci [i] := Ci[i]+1 As before, ticking is considered to occur before any other event action, and the timestamp C(e) of an event e is the clock value after ticking.
In "Scalability Approaches for Causal Multicast: A Survey" page 7 saysEach time a message m is multicast by pi , it is tagged with the local vector clock of its sender (i.e., VCm := VCpi ), once such vector clock has increased its local entry (i.e., VCpi [i] := VCpi [i] +1).
How to know if we can deliver a message?
(TODO: someones compare with <= and others say that is needed the +1 rule, see Scalability Approaches for Causal Multicast: A Survey)