You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If we add millisecond precision to ksuid then we will add to create a for and the size will increase to 29 bytes
The other option is we move to something like ULID .
In the previous issue related to ksuid we just solved that the current timestamp should be considered, but we ignored the precsion of the timestamp considerd
Yes, we can do that the size will be 32 bytes for that.
Do you think if we should add ns precision and also play around with the Random payload of 128 bits to reduce it down as we already have ns precision we just need something for randomness, or 32 bytes should be good ?
KSUID seems to have second precision and not ms/ns precision because of which id generation is shuffled for the same minute.
TS : 2024-11-04T18:02:07.026Z - id :
2oOayA9mfiFdFG4M1brPu3rfO0X
TS: 2024-11-04T18:02:07.025Z. - id:
2oOayC2SOU84wN95sZOIfMDMLDJ
The way to solve this will be ms/ns level presion in ksuid.
cc @srikanthccv
The text was updated successfully, but these errors were encountered: