raboof/pekko
1
0
Fork 0
Code Issues Pull requests Projects Releases 6 Packages Wiki Activity Actions 11

Merge pull request #23051 from manuelbernhardt/patch-1

Browse source 
Improving cluster specification documentation
This commit is contained in:
Patrik Nordwall 2017-06-12 10:23:18 +02:00 • committed by GitHub
commit 8241cd95f6
1 changed files with 6 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

11
akka-docs/src/main/paradox/scala/common/cluster.md
View file

@ -88,7 +88,7 @@ The failure detector is responsible for trying to detect if a node is
`unreachable` from the rest of the cluster. For this we are using an
implementation of [The Phi Accrual Failure Detector](http://www.jaist.ac.jp/~defago/files/pdf/IS_RR_2004_010.pdf) by Hayashibara et al.
An accrual failure detector decouple monitoring and interpretation. That makes
An accrual failure detector decouples monitoring and interpretation. That makes
them applicable to a wider area of scenarios and more adequate to build generic
failure detection services. The idea is that it is keeping a history of failure
statistics, calculated from heartbeats received from other nodes, and is
@ -176,11 +176,10 @@ seen set (have seen the new state) then the cluster gossips 3 times instead of o
every second. This adjusted gossip interval is a way to speed up the convergence process
in the early dissemination phase after a state change.
The choice of node to gossip with is random but it is biased to towards nodes that
might not have seen the current state version. During each round of gossip exchange when
no convergence it uses a probability of 0.8 (configurable) to gossip to a node not
part of the seen set, i.e. that probably has an older version of the state. Otherwise
gossip to any random live node.
The choice of node to gossip with is random but biased towards nodes that might not have seen
the current state version. During each round of gossip exchange, when convergence is not yet reached, a node
uses a very high probability (which is configurable) to gossip with another node which is not part of the seen set, i.e.
which is likely to have an older version of the state. Otherwise it gossips with any random live node.
This biased selection is a way to speed up the convergence process in the late dissemination
phase after a state change.