> On Aug 10, 2020, at 12:06 PM, Peter J. Holzer <hjp-pgsql@xxxxxx> wrote: > > On 2020-08-10 09:10:00 -0800, Israel Brewster wrote: >> I would point out, however, that using a V1 UUID rather than a V4 can >> help with this as it is sequential, not random (based on MAC address >> and timestamp + random). > > If I read the specs correctly, a V1 UUID will roll over every 429 > seconds. I think that as far as index locality is concerned, this is > essentially random for most applications. According to wikipedia, the time value in a V1 UUID is a 60-bit number, and will roll over "around 3400AD”, depending on the algorithm used, or 5236AD if the software treats the timestamp as unsigned. This timestamp is extended by a 13 or 14-bit “uniqifying" clock sequence to handle cases of overlap, and then the 48bit MAC address (constant, so no rollover there) is appended. So perhaps that 13 or 14 bit “uniqifying” sequence will roll over every 429 seconds, however the timestamp *as a whole* won’t roll over for quite a while yet, thereby guaranteeing that the UUIDs will be sequential, not random (since, last I checked, time was sequential). --- Israel Brewster Software Engineer Alaska Volcano Observatory Geophysical Institute - UAF 2156 Koyukuk Drive Fairbanks AK 99775-7320 Work: 907-474-5172 cell: 907-328-9145 > > hp > > -- > _ | Peter J. Holzer | Story must make more sense than reality. > |_|_) | | > | | | hjp@xxxxxx | -- Charles Stross, "Creative writing > __/ | http://www.hjp.at/ | challenge!"
