Method and apparatus for preprocessing technique for forecasting in capacity management, software rejuvenation and dynamic resource allocation applications
First Claim
1. In a computer system where forecasting of computing resources is performed based on past and present observations of measurements related to said resources, a computer-implemented method for preprocessing including decomposing said past and present observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence, the computer-implemented method comprising:
- detecting the spikes in a signal representing said measurements;
detecting the jumps in said signal;
removing spikes and jumps from said signal;
removing the noise from the signal, to obtain a smooth version of the signal, thereby to provide a basis for said forecasting of said computing resources by said computer system; and
providing, by said computer system, said forecast based on said smooth version of the signal.
1 Assignment
0 Petitions
Accused Products
Abstract
A system and method preprocesses data, in a computer system where forecasting of computing resources is performed based on past and present observations of measurements related to the resources. The preprocessing includes decomposing the past and present observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence. The method (and system) includes detecting the spikes in a signal representing the measurements, detecting the jumps in the signal, removing spikes and jumps from the signal, and removing the noise from the signal, to obtain a smooth version of the signal.
-
Citations
30 Claims
-
1. In a computer system where forecasting of computing resources is performed based on past and present observations of measurements related to said resources, a computer-implemented method for preprocessing including decomposing said past and present observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence, the computer-implemented method comprising:
-
detecting the spikes in a signal representing said measurements; detecting the jumps in said signal; removing spikes and jumps from said signal; removing the noise from the signal, to obtain a smooth version of the signal, thereby to provide a basis for said forecasting of said computing resources by said computer system; and providing, by said computer system, said forecast based on said smooth version of the signal. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A computer system, comprising:
-
a forecast module for forecasting computing resources based on observations of measurements related to said resources; and a data preprocessing module for decomposing said observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence, wherein said data preprocessing module comprises; an estimator for estimating a variance of the noise; a detector for detecting the spikes and jumps in said estimated variance of the noise; a subtractor for subtracting spikes and jumps from the signal, said estimator further estimating the variance of the noise of said signal based on said spikes and jumps having been removed, and said subtractor removing the noise, to obtain a smooth version of the signal, for input to said forecasting module to provide a basis for said forecasting of said computing resources by said forecasting module, wherein said forecasting module provides said forecasting based on said smooth version of the signal.
-
-
29. A signal-bearing medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a computer-implemented method for preprocessing data to be used for forecasting of computing resources based on observations of measurements related to said resources, said computer-implemented method for preprocessing including:
-
decomposing said observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence, said decomposing comprising; detecting the spikes in a signal representing said measurements; detecting the jumps in said signal; removing spikes and jumps from said signal; removing the noise from the signal, to obtain a smooth version of the signal, thereby to provide a basis for said forecasting of said computing resources by said digital data processing apparatus; and providing, by said digital data processing apparatus, said forecast based on said smooth version of the signal.
-
-
30. In a computer system where forecasting of computing resources is performed based on past and present observations of measurements related to said resources, a computer-implemented method for preprocessing including decomposing said past and present observations into a smooth time sequence, a jump time sequence, a noise time sequence and a spike time sequence, the computer-implemented method comprising:
-
detecting the spikes in a signal representing said measurements; detecting the jumps in said signal; removing spikes and jumps from said signal; removing the noise from the signal, to obtain a smooth version of the signal, thereby to provide a basis for said forecasting of said computing resources by said computer system; and providing, by said computer system, said forecast based on said smooth version of the signal, wherein at least one of said computing resources is exhaustible, and wherein the removing of the noise comprises; estimating the variance of the noise prior to said detecting of said spikes, wherein estimates of the quantities necessary to decompose the sequence are performed by first applying an invertible transform to the data associated with said observations and wherein said transform comprises a discrete wavelet transform (DWT), wherein the estimating the variance comprises estimating a variance of the noise using the coefficients of a highest frequency subband of the wavelet transform, and wherein the spikes are detected as up-and-down and down-and-up local variations, said spikes being judged not to be noise based on a result of a statistical test based on said estimating of the noise variance, wherein said forecasting is performed by estimating a trend of the smooth time sequence, and wherein seasonal components are separately removed from the smooth time sequence, the noise sequence, the spike sequence and the jump sequence, and wherein said forecasting is performed on each of said smooth time sequence, said jump time sequence, said noise time sequence, and said spike time sequence.
-
Specification