Timestamp is for ordering, and for determining decay of the strength factor. The ordering of events is not guaranteed by HDFS / Kafka, and does matter to some extent, especially if there are 'delete' events. It also matters when figuring out how old a data point is and how much its value has decayed, if it's enabled.   You could use seconds or milliseconds, I suppose, if you used them consistently. However the serving layer uses a standard ms timestamp, so that's probably best to emulate.     ... View more

Actually, I don't know the exact reasons and had stuck in this problem for a few day with firewalls on all machines disabled at very first. I used to deploy hadoop, spark and so on by extracting source tarballs. Forturnately, edge node seems to be a good idea to acess cluster resources. ... View more