Claims:1. A method for determining whether to admit a query in a network, the method comprising:
determining a load for a network element type based on an adaptive history for that network element type;
determining a cost of admitting the query based on the relative load that the query generates accounting for the amount of traffic the network element has admitted in the past;
decreasing a total cost of all queries that can be budgeted during a subsequent interval when the change in load is within a specified range;
increasing the total cost of all queries that can be budgeted during a subsequent interval when the change in load is below a threshold; and
adding the query to a data structure which keeps track of potentially admittable queries.

2. The method of claim 1 wherein the determining whether to admit a network query based on the adaptive history of the load is done by calculating the load according to:
LoadDL_NEi = (Average downlink Load of all network element of type i) + (n * si (Standard Deviation) of downlink load of all network element of type i), where LoadDL_NEi is the representative downlink load of the network equipment of type i, for iteration i, at the beginning of the interval and n is a positive number that permits fine-tuning of the aggressiveness of the algorithm.

3. The method of claim 2 wherein determining the cost of admitting the query based on the relative load that the query generates is done by calculating the cost according to:
Cost_AdmCUR_DL_NEi = sum of CostQDL of all mission-critical queries admitted, where Cost_AdmCUR_DL_NEi is the sum of CostQDL of all queries that may be potentially admitted during the current time interval, considering the load of network element of type i in the downlink direction, and CostQDL = (WDL1 * DL Data volume of the query )* (WDL2(number of devices to be queried, number of network elements of the given type)), where WDL1 and WDL2 are weighting factors and WDL2 is a function f(number of devices to be queried, number of network elements of the given type).

4. The method of claim 3 wherein the change in load is within a range when:
?DL_Th1 < ?DL_Load_NEi < ?DL_Max, ?DL_Load_NEi = (downlink load at the end of last (nth) interval – downlink load at the beginning of the first interval) / n. ? DL_Max and ?DL_Th1 are thresholds which are functions of LoadDL_NEi.

5. The method of claim 4 wherein decreasing the total cost of all queries that can be budgeted is done by assigning Xlast_DL_NEi = Xlast_DL_NEi / FactorDL1, where Xlast_DL_NEi is the total cost of all queries that can be budgeted during the subsequent interval and FactorDL1, is a number between 1 and 2.

6. The method of claim 5 wherein increasing the total cost of all queries that can be budgeted during a subsequent interval occurs when ?DL_Load_NEi < ?DL_Th2; where ?DL_Th2 is a threshold which is a function of LoadDL_NEi.

7. The method of claim 6 wherein increasing the total cost of all queries that can be budgeted occurs by assigning Xlast_DL_NEi = minimum((Xlast_DL_NEi * FactorDL2), maximum(Xlast_DL_NEi, (Cost_Admlast_DL_NEi * FactorDL2))), where FactorDL2, is a number between 1 and 2.

8. The method of claim 7 wherein in an initial interval after calculating load and cost, assigning Xlast_DL_NEi = XInit_DL_NEi, where XInit_DL_NEi represents the total cost of all queries that can be budgeted during the initial interval and is a function f(LoadDL_NEi, total capacity of network element of type i).

9. The method of claim 1 further comprising, selecting the set of network element queries which are allowed in all types of network elements for consideration of admission.

10. The method of claim 1 wherein all mission critical queries are admitted.

11. The method of claim 1 wherein queries to be admitted on all types of NEs are selected based on information in the data structure.

12. The method of claim 2 wherein comparing the downlink load of the network element type i to the set threshold and comparing the change in the downlink load of the network element type i, averaged over the last n intervals, to a second threshold are done using comparisons according to the following expressions respectively:
((LoadDL_NEi >=ThreshHigh_DL) OR (?DL_Load_NEi >=? DL_Max).

13. The method of claim 1 further comprising:
for each type of NE (i = 1 to N), calculating Cost_Admlast_DL_NEi as the sum of CostQDL of the finally selected set of admitted queries;
for each type of NE (i = 1 to N), calculating Cost_Admlast_UL_NEi as the sum of CostQUL of the finally selected set of admitted queries; and
selecting a set of NE queries that are allowed based on the Cost_Admlast_UL_NEi and Cost_Admlast_DL_NEi.

14. The method of claim 1 wherein the determining whether to admit a network query based on the adaptive history of the load is done by calculating the load according to:
LoadUL_NEi = (Average uplink Load of all network element of type i) + (n * si (Standard Deviation) of uplink load of all network element of type i), where LoadUL_NEi is the representative uplink load of the network equipment of type i, for iteration i, at the beginning of the interval and n is a positive number that permits fine-tuning of the aggressiveness of the algorithm.

15. A device for determining whether to admit a query in a network, the device comprising:
a memory; and
a processor configured to:
determine a load for a network element type based on an adaptive history for that network element type;
determine a cost of admitting the query based on the relative load that the query generates accounting for the amount of traffic the network element has admitted in the past;
decrease a total cost of all queries that can be budgeted during a subsequent interval when the change in load is within a specified range;
increase the total cost of all queries that can be budgeted during a subsequent interval when the change in load is below a threshold; and
add the query to a data structure which keeps track of potentially admittable queries.

16. The device of claim 15 wherein the processor is further configured to select the set of network element queries which are allowed in all types of network elements for consideration of admission.

17. The device of claim 15 wherein all mission critical queries are admitted.

18. A data storage device encoding a program in non-transitory machine readable and machine-executable form, the program comprising instructions to:
determine a load for a network element type based on an adaptive history for that network element type;
determine a cost of admitting the query based on the relative load that the query generates accounting for the amount of traffic the network element has admitted in the past;
decrease a total cost of all queries that can be budgeted during a subsequent interval when the change in load is within a specified range;
increase the total cost of all queries that can be budgeted during a subsequent interval when the change in load is below a threshold; and
add the query to a data structure which keeps track of potentially admittable queries.

19. The data storage device of claim 18 further comprising instructions to select the set of network element queries which are allowed in all types of network elements for consideration of admission.

20. The data storage device of claim 18 wherein all mission critical queries are admitted.
, Description:N/A