Dynamic resource partitioning for long-term fairness to non-elastic traffic on a cellular basestation
View Patent ↗A method includes receiving non-elastic traffic into a cellular network serving users; varying long-term dissatisfaction of the users with dynamic resource allocations during traffic overload of the cellular network, the long term dissatisfaction being measured by at least one of a function of number of times a user from the users receives bad quality of service and a level of deterioration of quality of service to the user; and influencing number of satisfied users of the cellular network responsive to the varying of long-term dissatisfaction.
1. A method comprising:
receiving, at a base station, non-elastic traffic into a cellular network serving users;
varying long-term dissatisfaction of said users with dynamic resource allocations during traffic overload of said cellular network, said long term dissatisfaction being measured by at least one of a function of a number of times a user from said users receives a bad quality of service and a level of deterioration of a quality of service to said user; and
influencing a number of satisfied said users of said cellular network responsive to said varying long-term dissatisfaction,
wherein said varying comprises separating flows dynamically into two groups or slices, and migrating, upon detection of overload in a high priority slice, flows with a minimum long-term dissatisfaction to a low priority slice, and wherein said flows are migrated to the low priority slice in an increasing order of the long-term dissatisfaction and to the high priority slice after a configurable amount of time DELTA.
2. The method of claim 1 , wherein said level of deterioration is a function of a number of bytes requested over said cellular network, the number of bytes reserved by flow on said cellular network, and the number of bytes actually served over said cellular network.
3. The method of claim 1 , wherein said varying comprises giving flows in a first slice higher priority over flows in a second slice.
4. The method of claim 1 , wherein said varying comprises initially giving flows higher priority and initially servicing said flows using a first-come-first-served or earliest deadline first policy.
5. The method of claim 1 , wherein said overload is detected when a cumulative arrival rate of traffic of all flows exceeds a service rate of said network by a predetermined threshold.
6. The method of claim 1 , wherein said flows in the low priority slice are served using a shortest queue first policy.
7. The method of claim 6 , wherein if marked or detectable, packets of the flows in the low priority slice are dropped in favor of packets of the flows in the high priority slice.
8. A method comprising:
receiving, at a base station, non-elastic traffic into a cellular network serving users;
influencing long-term dissatisfaction of said users on said cellular network by allocating information flows over said cellular network during traffic overload of said cellular network, said dissatisfaction being measured by at least one of a function of a number of times a user from said users receives a bad quality of service and a level of deterioration of a quality of service to said user; and
changing a number of satisfied said users of said cellular network responsive to said influencing,
wherein said influencing comprises separating flows dynamically into two groups or slices, and migrating, upon detection of overload in a high priority slice, flows with a minimum long-term dissatisfaction to a low priority slice, and
wherein said flows are migrated to the low priority slice in an increasing order of the long-term dissatisfaction and to the high priority slice after a configurable amount of time DELTA.