Actions
Feature #1973
openVLR: efficient subscriber lookup
Status:
New
Priority:
Low
Assignee:
-
Category:
-
Target version:
-
Start date:
03/08/2017
Due date:
% Done:
0%
Resolution:
Spec Reference:
Description
In libvlr, we (will) keep every subscriber in RAM as long as it still has valid auth tuples (3GPP TS 33.102 Annex C.2.3).
Most subscribers will still have auth tuples left upon detaching, so we would store all subscribers.
Storing numerous subscribers in a linear list is inefficient, implement a hash table for faster access times.
Related issues
Updated by neels about 7 years ago
- Related to Feature #1974: VLR: high water mark on subscriber storage added
Updated by neels about 7 years ago
- Subject changed from VLR: efficient subscriber storage to VLR: efficient subscriber lookup
Actions