System and method for statistical performance monitoring
First Claim
Patent Images
1. A method for reducing the amount of data of system metrics collected or reported from agent nodes to a system performance monitor for system performance monitoring and analysis, the method comprising the steps of:
- obtaining a sampled value of a first system metric;
reporting the sampled value of the first system metric if the sampled value is not between a first parameter and a second parameter, wherein the first parameter and the second parameter are any real numbers;
not reporting the sampled value if the sampled value is between the first and second parameters;
calculating a weighted running average, wherein
{overscore (d)}n(w)=dnw+{overscore (d)}n−
1(1−
w),{overscore (d)}n and {overscore (d)}n−
1 are the weighted running average after n'"'"'th or (n−
1)'"'"'th sampling,w is the weighing factor for the sampling,
Sn=Sn−
1+(n−
1)(dn−
{overscore (d)}n−
1)2/n,
σ
n2=Sn/n,Sn and Sn−
1 are the sum of the differences squared,σ
n is the standard deviation,calculating the first parameter to be ({overscore (d)}n−
aσ
n); and
calculating the second parameter to be ({overscore (d)}n+bσ
n), wherein a and b are two constant real numbers.
9 Assignments
0 Petitions
Accused Products
Abstract
A method using statistical parameters (e.g. mean, standard deviation, exceptional values) of performance monitoring metrics to substantially reduce the quantity of performance monitoring data collected and reported, make system performance monitoring scalable and enhance the readability of the system performance display. The number of metrics monitored may be reduced by monitoring only one of any two metrics that are closely correlated.
53 Citations
57 Claims
-
1. A method for reducing the amount of data of system metrics collected or reported from agent nodes to a system performance monitor for system performance monitoring and analysis, the method comprising the steps of:
-
obtaining a sampled value of a first system metric; reporting the sampled value of the first system metric if the sampled value is not between a first parameter and a second parameter, wherein the first parameter and the second parameter are any real numbers; not reporting the sampled value if the sampled value is between the first and second parameters; calculating a weighted running average, wherein
{overscore (d)}n(w)=dnw+{overscore (d)}n−
1(1−
w),{overscore (d)}n and {overscore (d)}n−
1 are the weighted running average after n'"'"'th or (n−
1)'"'"'th sampling,w is the weighing factor for the sampling,
Sn=Sn−
1+(n−
1)(dn−
{overscore (d)}n−
1)2/n,
σ
n2=Sn/n,Sn and Sn−
1 are the sum of the differences squared,σ
n is the standard deviation,calculating the first parameter to be ({overscore (d)}n−
aσ
n); andcalculating the second parameter to be ({overscore (d)}n+bσ
n), wherein a and b are two constant real numbers. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A computer system module for system performance monitoring, reporting and analysis, the module comprising:
-
a controller module operative to control the system performance monitoring; and a sampling module coupled to the controller module, operative to sample at least a first system metric, and obtain a sampled value of the first system metric, wherein each sampled value of the first system metric is reported if the sampled value is not between a first parameter and a second parameter, and not reported if the sampled value is between the first and second parameters, wherein the first parameter and the second parameter are any real numbers, and wherein the controller module is operative to calculate a weighted running avenge, wherein
{overscore (d)}n(w)=dnw+{overscore (d)}n−
1(1−
w),{overscore (d)}n and {overscore (d)}n−
1 are the weighted running average after n'"'"'th or (n−
1)'"'"'th sampling,w is the weighing factor for the sampling,
Sn=Sn−
1+(n−
1)(dn−
{overscore (d)}n−
1)2/n,
σ
n2=Sn/n,Sn and Sn−
1 are the sum of the differences squared, andσ
n is the standard deviation; andcalculate the first parameter to be ({overscore (d)}n−
aσ
n) and the second parameter to be ({overscore (d)}n+bσ
n), wherein a and b are two constant real numbers. - View Dependent Claims (22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37)
-
-
38. A computer network system comprising:
-
a plurality of network nodes having a CPU; a memory module coupled to CPU, operative to contain computer executable programs; and a network interface operative to interconnect different nodes of the network, wherein one computer executable program is loaded in the memory module in one node, wherein the computer executable program is operative to perform a method for reducing the amount of data of system metrics collected or reported from agent nodes to a system performance monitor for system performance monitoring and analysis, the method comprising the steps of; obtaining a sampled value of a first system metric; reporting the sampled value of the first system metric if the sampled value is not between a first parameter and a second parameter, wherein the first parameter and the second parameter are any real numbers; not reporting the sampled value if the sampled value is between the first and second parameters; calculating a weighted running average, wherein
{overscore (d)}n(w)=dnw+{overscore (d)}n−
1(1−
w),{overscore (d)}n and {overscore (d)}n−
1 are the weighted running average after n'"'"'th or (n−
1)'"'"'th sampling,w is the weighing factor for the sampling,
Sn=Sn−
1+(n−
1)(dn−
{overscore (d)}n−
1)2/n,
σ
n2=Sn/n,Sn and Sn−
1 are the sum of the differences squared,σ
n is the standard deviation,calculating the first parameter to be ({overscore (d)}n−
aσ
n); andcalculating the second parameter to be ({overscore (d)}n+bσ
n), wherein a and b are two constant real numbers. - View Dependent Claims (39, 40, 41, 42, 43, 44, 45, 46, 47)
-
-
48. A machine readable medium comprising a machine executable program, wherein the machine executable program is operative to perform a method for reducing the amount of data of system metrics collected or reported from agent nodes to a system performance monitor for system performance monitoring and analysis, the method comprising the steps of:
-
obtaining a sampled value of a first system metric; reporting the sampled value of the first system metric if the sampled value is not between a first parameter and a second parameter, wherein the first parameter and the second parameter are any real numbers; not reporting the sampled value if the sampled value is between the first and second parameters; calculating a weighted running average, wherein
{overscore (d)}n(w)=dnw+{overscore (d)}n−
1(1−
w),{overscore (d)}n and {overscore (d)}n−
1 are the weighted running average after n'"'"'th or (n−
1)'"'"'th sampling,w is the weighing factor for the sampling,
Sn=Sn−
1+(n−
1)(dn−
{overscore (d)}n−
1)2/n,
σ
n2=Sn/n,Sn and Sn−
1 are the sum of the differences squared,σ
n is the standard deviation,calculating the first parameter to be ({overscore (d)}n−
aσ
n); andcalculating the second parameter to be ({overscore (d)}n+bσ
n), wherein a and b are two constant real numbers. - View Dependent Claims (49, 50, 51, 52, 53, 54, 55, 56, 57)
-
Specification