Current server team collection logic does not consider
the fact that multipe storage servers can run on the same machine.
When multiple machines fail, all servers on the machines will fail, and
the possibility of having one process team fail and lose data is very high.
To reduce the possibility of losing data when multiple machine fails,
we first create machine teams which span across different fault zones;
we then create server teams based on machine teams by
first picking 1 machine team, and then
picking 1 server from each machine in the machine team.
Signed-off-by: Meng Xu <meng_xu@apple.com>
Remove the use of relative paths. A header at foo/bar.h could be included by
files under foo/ with "bar.h", but would be included everywhere else as
"foo/bar.h". Adjust so that every include references such a header with the
latter form.
Signed-off-by: Robert Escriva <rescriva@dropbox.com>
Construction of LocalityGroup from LocalityData is expensive, and the previous
code greatly ran afoul of that. The policy engine does a large amount of
interning of strings and building compressed maps to make the expected many
future selectReplica calls cheap. Unfortunately we don't call selectReplicas,
so much of this work is undesireable for us, and a large amount of CPU time is
spent doing this initialization work.
The new changes aggressively do the minimal LocalityGroup::add() calls
necessary, and make them as cheap as possibly by removing all elements from
LocalityData that don't need to be considered by the policy.
This optimization was also applied to the PeekCursor used during recovery,
which should speed recoveries up by a small amount.
A.J. Beamon