An interesting and well-written paper.<p>The authors not only advocate remarkably well their approach
for protocol-aware recovery of distributed systems,
they also give a broad overview of storage faults, their adverse impacts on consensus protocols and the possible recovery strategies.