commit 602907050b948ecba87b30bff8cfda0e5f91ce95
parent a1250219ecda0aa0dcc82f6cc821bc889f28b54c
Author: Ivan Gankevich <igankevich@ya.ru>
Date: Mon, 2 Oct 2017 10:12:57 +0300
Update evaluation section.
Diffstat:
1 file changed, 12 insertions(+), 7 deletions(-)
diff --git a/arma-thesis.org b/arma-thesis.org
@@ -2982,19 +2982,24 @@ cluster.
Node discovery performance was evaluated by measuring time needed for all nodes
of the cluster to discover each other, i.e. the time needed for the tree
hierarchy of nodes to reach stable state. Each change of the hierarchy, as seen
-by each node, was written to a log file and after 30 seconds all daemon
-processes (each of which models cluster node) were forcibly terminated. Each new
-daemon process was launched with a 100ms delay to ensure that principal nodes
-are always come online before subordinates and hierarchy does not change
+by each node, was written to a log file and after a set amount of time all
+daemon processes (each of which models cluster node) were forcibly terminated.
+Each new daemon process was launched with a 100ms delay to ensure that principal
+nodes are always come online before subordinates and hierarchy does not change
randomly as a result of different start time of each process. This artificial
delay was subsequently subtracted from the results. So, benchmark results
represent discovery time in an "ideal" cluster, in which every daemon always
finds its principal on the first try.
The benchmark was run varying the number of daemons per cluster node. The
-experiment showed that discovery of up to 400 nodes each other takes no more
-than 2 seconds (fig.\nbsp{}[[fig-discovery-benchmark]]). This value does not change
-significantly with the increase in number of physical nodes.
+experiment showed that discovery of up to 512 (8 nodes with 64 processes per
+node) nodes each other takes no more than 2 seconds
+(fig.\nbsp{}[[fig-discovery-benchmark]]). This value does not change significantly
+with the increase in the number of physical nodes. Using more than 8 nodes with
+64 processes per node causes large variation in discovery time due to a large
+total number of processes connecting simultaneously to one master process (a
+fan-out value of 10000 was used for all tests), so these results were excluded
+from the benchmark.
#+name: fig-discovery-benchmark
#+header: :width 8 :height 6
