LogDevice has many-to-many rebuilding as well, and typically data for a log (similar to partition in Kafka) is spread relatively uniformly over the (potentially

LogDevice has many-to-many rebuilding as well, and typically data for a log (similar to partition in Kafka) is spread relatively uniformly over the (potentially large, much bigger than replication factor) set of shards that hold data for that log.