Question: Understanding Units in Trace Data

[haomactracy]

Hi， i want to ask 2 question， 1. in trace.csv, the unit of every cols is version, time(ms or s ?), op, size(byte), lbn(512B,sector)? 2. struct { uint32_t timestamp; uint64_t obj_id; uint32_t obj_size; int64_t next_access_vtime; // -1 if no next access } what are the units of timestamp and next_access_vtime, if we have timestamp, why we need next_acess_vtime

thanks

[haomactracy]

here are strips in trace.csv 1,5633898,2a,512,42932745 1,5633898,2a,512,42932746 1,5633898,2a,512,42932747

[haomactracy]

the second question is for twitter_cluster52.csv

what are their units time, object(?), size(byte), next_access_vtime(ms) 0, 13053225291711363978, 737, 13 0, 61177148907475485, 248, 6570713 0, 286355432205858634, 567, 7 0, 15247269157424090910, 54, 1244 0, 4675217417842824703, 54, 1650187 0, 6006438795856014400, 253, 7142982

[1a1a11a]

1. the trace.csv time is in sec, the next_access_time is the logical time (# requests) between current and the next request. It is used for some algorithms that require future information, e.g., Belady
2. object id is a hash of object id (string)