Adaptive high efficiency, low latency DOCSIS PGS scheduler
Determining when to provide a Proactive Grant Service (PGS) scheduling grant. A plurality of PGS grants are issued to a cable modem (CM). The PGS grants that were utilized by the CM are monitored as well as those PGS grants that were not utilized by the CM. A compromise PGS grants pattern for that CM is generated based on the observations of which PGS grants the CM utilized and which PGS grants the CM did not utilize. The compromise PGS grants pattern for that CM optimizes a projected experienced latency and jitter for particular data flows of the CM verses a projected wasted upstream capacity.
1. A non-transitory computer-readable storage medium storing one or more sequences for determining when to provide a Proactive Grant Service (PGS) scheduling grant, which when executed by one or more processors, cause:
monitoring which PGS grants, of a plurality of PGS grants issued to a cable modem (CM), were utilized by the CM and which PGS grants were not utilized by the CM; and
generating, based on observations of when PGS grants in the plurality of PGS grants were utilized by said CM and when PGS grants in the plurality of PGS grants were not utilized by said CM, a compromise PGS grants pattern of one or more regular intervals of time at which to issue a PGS scheduling grant for that CM to optimize a projected experienced latency and jitter for particular data flows verses a projected wasted upstream capacity, wherein each of said one or more regular intervals of time is located at a specified point in time within said compromise PGS grants pattern.
2. The non-transitory computer-readable storage medium of claim 1 , wherein generating the compromise PGS grants pattern results in less wasted US capacity as the total demand for the available DOCSIS US capacity approaches full utilization.
3. The non-transitory computer-readable storage medium of claim 1 , wherein generating the compromise PGS grants pattern results in lower latency and jitter for said CM's particular data flows while the total demand for the available DOCSIS US capacity is less than a determined amount.
4. The non-transitory computer-readable storage medium of claim 1 , wherein generating the compromise PGS rate is performed using a set of optimization weights.
5. The non-transitory computer-readable storage medium of claim 1 , wherein execution of the one or more sequences of instructions further causes:
repeating the generation of said compromise PGS grants pattern at periodic intervals or upon determining a change in a condition has occurred.
6. The non-transitory computer-readable storage medium of claim 1 , wherein said generating the compromise PGS grants pattern comprises:
refining the compromise PGS grants pattern for the CM over time using measurements obtained at different points in time.
7. The non-transitory computer-readable storage medium of claim 1 , wherein said generating the compromise PGS grants pattern comprises:
arranging a group of grants such that a time interval between consecutive grants is less than one bandwidth allocation MAP time.
8. The non-transitory computer-readable storage medium of claim 1 , wherein said generating the compromise PGS grants pattern comprises:
performing a first set of grants at a fixed time interval for a particular flow; and
performing an occasional additional grant at a variable time interval for the particular flow to detect a possible timing drift in the particular flow.
9. The non-transitory computer-readable storage medium of claim 1 , wherein the compromise PGS grants pattern is optimized for one or more of: expected periodicity, expected time of arrival for each data packet, and expected data packet size of the particular data flows of the CM.
10. An apparatus for determining when to provide a Proactive Grant Service (PGS) scheduling grant, comprising:
one or more processors; and
one or more non-transitory computer-readable storage mediums storing one or more sequences of instructions, which when executed, cause:
monitoring which PGS grants, of a plurality of PGS grants issued to a cable modem (CM), were utilized by the CM and which PGS grants were not utilized by the CM; and
generating, based on observations of when PGS grants in the plurality of PGS grants were utilized by said CM and when PGS grants in the plurality of PGS grants were not utilized by said CM, a compromise PGS grants pattern of one or more regular intervals in time at which to issue a PGS scheduling grant for that CM to optimize a projected experienced latency and jitter for particular data flows verses a projected wasted upstream capacity, wherein each of said one or more regular intervals of time is located at a specified point in time within said compromise PGS grants pattern.
11. The apparatus of claim 10 , wherein generating the compromise PGS grants pattern results in less wasted US capacity as the total demand for the available DOCSIS US capacity approaches full utilization.
12. The apparatus of claim 10 , wherein generating the compromise PGS grants pattern results in lower latency and jitter for said CM's particular data flows while the total demand for the available DOCSIS US capacity is less than a determined amount.
13. The apparatus of claim 10 , wherein generating the compromise PGS rate is performed using a set of optimization weights.
14. The apparatus of claim 10 , wherein execution of the one or more sequences of instructions further causes:
repeating the generation of said compromise PGS grants pattern at periodic intervals or upon determining a change in a condition has occurred.
15. The apparatus of claim 10 , wherein said generating the compromise PGS grants pattern comprises:
refining the compromise PGS grants pattern for the CM over time using measurements obtained at different points in time.
16. The apparatus of claim 10 , wherein said generating the compromise PGS grants pattern comprises:
arranging a group of grants such that a time interval between consecutive grants is less than one bandwidth allocation MAP time.
17. The apparatus of claim 10 , wherein said generating the compromise PGS grants pattern comprises:
performing a first set of grants at a fixed time interval for a particular flow; and
performing an occasional additional grant at a variable time interval for the particular flow to detect a possible timing drift in the particular flow.
18. The apparatus of claim 10 , wherein the compromise PGS grants pattern is optimized for one or more of: expected periodicity, expected time of arrival for each data packet, and expected data packet size of the particular data flows of the CM.
19. A method for determining when to provide a Proactive Grant Service (PGS) scheduling grant, comprising:
monitoring which PGS grants, of a plurality of PGS grants issued to a cable modem (CM), were utilized by the CM and which PGS grants were not utilized by the CM; and
generating, based on observations of when PGS grants in the plurality of PGS grants were utilized by said CM and when PGS grants in the plurality of PGS grants were not utilized by said CM, a compromise PGS grants pattern of one or more regular intervals in time at which to issue a PGS scheduling grant for that CM to optimize a projected experienced latency and jitter for particular data flows verses a projected wasted upstream capacity, wherein each of said one or more regular intervals of time is located at a specified point in time within said compromise PGS grants pattern.
20. The method of claim 19 , wherein generating the compromise PGS grants pattern results in less wasted US capacity as the total demand for the available DOCSIS US capacity approaches full utilization.
21. The method of claim 19 , wherein generating the compromise PGS grants pattern results in lower latency and jitter for said CM's particular data flows while the total demand for the available DOCSIS US capacity is less than a determined amount.
22. The method of claim 19 , wherein generating the compromise PGS rate is performed using a set of optimization weights.
23. The method of claim 19 , further comprising:
repeating the generation of said compromise PGS grants pattern at periodic intervals or upon determining a change in a condition has occurred.
24. The method of claim 19 , wherein said generating the compromise PGS grants pattern comprises:
refining the compromise PGS grants pattern for the CM over time using measurements obtained at different points in time.
25. The method of claim 19 , wherein said generating the compromise PGS grants pattern comprises:
arranging a group of grants such that a time interval between consecutive grants is less than one bandwidth allocation MAP time.
26. The method of claim 19 , wherein said generating the compromise PGS grants pattern comprises:
performing a first set of grants at a fixed time interval for a particular flow; and
performing an occasional additional grant at a variable time interval for the particular flow to detect a possible timing drift in the particular flow.
27. The method of claim 19 , wherein the compromise PGS grants pattern is optimized for one or more of: expected periodicity, expected time of arrival for each data packet, and expected data packet size of the particular data flows of the CM.