Management for a heterogeneous pool of processors for the assignment of additional load
View Patent ↗An exemplary method implements load management for large granularity processes on application processors, APs. First data associated with the primary processes running on each AP is periodically collected, where the first data is proportional to processor occupancy, PO, for the primary processes running on each AP. Second data associated with auxiliary processes running on each AP is periodically collected where the auxiliary processes directly support the primary processes running on the respective AP. The second data is proportional to PO for the auxiliary processes running on each AP. A processor scaling factor and an overhead scaling factor are calculated for each AP based on the first and second data, respectively. The total amount of additional PO a second AP would incur to run a first large granularity process is determined by two aspects. The amount of additional PO due to the primary process is determined by applying at least the second processor scaling factor to a value related to an amount of primary process PO of the first process running on the first AP. The amount of additional PO due to overhead processes is determined by applying the overhead scaling factor of the second AP to the previously determined amount of additional PO due to the primary processes determined for the second AP.
1. A method for implementing load management for large granularity processes on application processors (APs) comprising the steps of:
collecting first data associated with primary large granularity processes running on each AP, wherein the first data is proportional to processor occupancy (PO) for the primary processes running on each AP;
collecting second data associated with auxiliary processes running on each AP, wherein the auxiliary processes directly support the primary processes running on the respective AP, and the second data is proportional to PO for the auxiliary processes running on each AP;
calculating for each AP a processor scaling factor and an overhead scaling factor based on the first and second data, respectively; and
for a first large granularity process running on a first AP, determining a total amount of additional PO a second AP would incur to run the first process, wherein said determination comprises calculating a first amount by applying at least the second processor scaling factor to a value related to an amount of PO of the first process running on the first AP, and applying the second overhead scaling factor to the first calculated amount to reflect additional PO due to auxiliary processes.
2. The method of claim 1 wherein the first and second data are periodically collected over a time interval of sufficient length to produce a high probability of having encountered periods of maximum loading.
3. The method of claim 1 further comprising the step of storing for each AP a corresponding processor scaling factor and overhead scaling factor.
4. The method of claim 3 further comprising the step of updating the stored processor scaling factor and overhead scaling factor for each AP based on newly collected periodic first and second data, respectively.
5. The method of claim 1 wherein a large granularity process is defined by a process resulting in a corresponding PO on an AP of 3%-30%.
6. The method of claim 1 wherein each of the steps is automatically implemented by an apparatus utilizing a microprocessor.
7. The method of claim 1 wherein the step of determining comprises the total amount determined by applying a ratio of the first and second processor scaling factors to a value related to an amount of primary process PO of the first process running on the first AP to calculate a first amount, and applying the second overhead scaling factor to the first amount to reflect additional PO due to auxiliary processes.
8. The method of claim 1 wherein the the step of determining comprises the total amount determined by multiplying the second processor scaling factor times a value related to an amount of primary process PO of the first process running on the first AP to calculate a first amount, and applying the second overhead scaling factor to the first amount to reflect additional PO due to auxiliary processes.
9. A method implemented by a computing system for determining load management for large granularity telecommunication processes on telecommunication application processors, APs, comprising the steps of:
collecting first data associated with primary large granularity telecommunication processes running on each telecommunication AP, where the first data is proportional to processor occupancy, PO, for the primary processes running on each AP;
collecting second data associated with auxiliary processes running on each telecommunication AP where the auxiliary processes directly support the primary telecommunication processes running on the respective AP, where the second data is proportional to PO for the auxiliary processes running on each AP;
calculating for each telecommunication AP a processor scaling factor and an overhead scaling factor based on the first and second data, respectively;
determining a total amount of additional PO a second telecommunication AP would incur to run a first large granularity process where the first process is running on a first telecommunication AP, where the total amount includes a first amount, the first amount of additional PO due to the primary large granularity processes being determined by applying at least the second processor scaling factor to a value related to an amount of primary process PO of the first process running on the first AP, and the total amount determined by applying the second overhead scaling factor to the first amount to reflect additional PO due to auxiliary processes.
10. The method of claim 9 wherein the first and second data are periodically collected over a time interval of sufficient length to produce a high probability of having encountered periods of maximum loading.
11. The method of claim 9 wherein the the step of determining comprises the first amount determined by multiplying the second processor scaling factor times a value related to an amount of primary process PO of the first process running on the first AP to calculate a first amount.
12. The method of claim 11 wherein the value is a number of messages associated by the primary process running on the first AP.
13. The method of claim 9 wherein a large regularity process is defined by a process resulting in a corresponding PO on an AP of 3%-30%.
14. The method of claim 9 wherein wherein the step of determining comprises the first amount determined by applying a ratio of the first and second processor scaling factors to the value related to the amount of primary process PO of the first process running on the first AP to calculate a first amount.
15. The method of claim 9 wherein the first data comprises a count of the number of messages generated by a telecommunication AP during a predetermined time interval.
16. A computing apparatus adapted for determining load management for large granularity processes on application processors, APs, comprising:
means for periodically collecting first data associated with primary large granularity processes running on each AP, where the first data is proportional to processor occupancy, PO, for the primary processes running on each AP;
means for periodically collecting second data associated with auxiliary processes running on each AP where the auxiliary processes directly support the primary processes running on the respective AP, where the second data is proportional to PO for the auxiliary processes running on each AP;
means for calculating for each AP a processor scaling factor and an overhead scaling factor based on the first and second data, respectively;
means for determining a total amount of additional PO a second AP would incur to run a first large granularity process where the first process is running on a first AP, where the total amount is determined by applying at least the second processor scaling factor to value related to n amount of primary process PO of the first process running on the first AP to calculate a first amount, and applying the second overhead scaling factor to the first amount to reflect additional PO due to auxiliary processes.
17. The computing apparatus of claim 16 further comprising means for storing a corresponding processor scaling factor and overhead scaling factor for each AP, and means for updating the stored processor scaling factor and overhead scaling factor for each AP based on newly collected periodic first and second data, respectively.
18. The computing apparatus of claim 16 wherein a large regularity process is defined by a process resulting in a corresponding PO on an AP of 3%-30%.
19. The computing apparatus of claim 16 wherein the means for periodically collecting the first data comprises means for counting of the number of messages generated by an AP during a predetermined time interval.
20. The computing apparatus of claim 16 wherein the large granularity processes comprise large granularity telecommunication processes run on telecommunication application processors.