I need to have text identifiers in my application. For example, we have an acceptable encoding azAZ09 and a valid range of identifiers [aaa] - [cZ9]. The first generated identifier will be aaa, then aab, aac, aad etc
How can I return the identifier and increase the lower border of the transaction? (provided that there are hundreds of simultaneous requests, and everyone should have the correct result)
To reduce the load, I assume that you can define 20 separate ranges and return the identifier from an arbitrary range - this should reduce competition, but it is not clear how to do one operation in the first place.
Also, note that the number of identifiers in a range can exceed 2 ^ 32.
Another idea is to have ranges of 64-bit integers and convert integer → char id to program code, where this could be done asynchronously.
Any ideas?
source
share