Hybrid event prediction and system control
First Claim
1. An information processing system for predicting an occurrence of a critical event in a computer cluster, the information processing system comprising:
- the computer cluster comprising a series of nodes;
a control system comprising;
A) an event log comprising information concerning critical events that occur in the computer cluster, wherein said critical events adversely affect performance of the cluster or one of its nodes;
B) a system parameter log comprising information concerning system performance parameters for each node in the cluster;
C) a processor configured for;
i) implementing a hybrid prediction system comprising rule based prediction algorithms, time-dependent variable prediction algorithms, and a warning window;
ii) maintaining the event log for only those nodes in which an error has occurred;
iii) maintaining the system parameter log for each node in the cluster;
iv) for only those nodes in which an error has occurred, loading the information from the event log and the system performance information pertaining to said error-prone nodes from the system parameter log into a Bayesian network model representing a correspondence between the system performance parameters and occurrence of the critical events;
v) using the Bayesian network model to predict a future critical event within a specified time-limit based upon the hybrid prediction system;
vi) making future scheduling and current data migration selections based upon the hybrid prediction system; and
vii) adapting the Bayesian Network Model by feeding the scheduling and data migration selections into said Bayesian Network Model; and
D) a memory for storing information related to said occurrence of said critical events and said system parameters of said nodes.
5 Assignments
0 Petitions
Accused Products
Abstract
A system for predicting an occurrence of a critical even in a computer cluster includes: a control system that includes an event log, a system parameter log, a memory for storing information related to occurrences of critical events, and a processor. The processor implements a hybrid prediction system; loads the information from the event log and the system performance log into a Bayesian network model; uses the Bayesian network model to predict a future critical event; makes future scheduling and current data migration selections; and adapts the Bayesian network model by feeding the scheduling and data migration selections.
28 Citations
18 Claims
-
1. An information processing system for predicting an occurrence of a critical event in a computer cluster, the information processing system comprising:
-
the computer cluster comprising a series of nodes; a control system comprising; A) an event log comprising information concerning critical events that occur in the computer cluster, wherein said critical events adversely affect performance of the cluster or one of its nodes; B) a system parameter log comprising information concerning system performance parameters for each node in the cluster; C) a processor configured for; i) implementing a hybrid prediction system comprising rule based prediction algorithms, time-dependent variable prediction algorithms, and a warning window; ii) maintaining the event log for only those nodes in which an error has occurred; iii) maintaining the system parameter log for each node in the cluster; iv) for only those nodes in which an error has occurred, loading the information from the event log and the system performance information pertaining to said error-prone nodes from the system parameter log into a Bayesian network model representing a correspondence between the system performance parameters and occurrence of the critical events; v) using the Bayesian network model to predict a future critical event within a specified time-limit based upon the hybrid prediction system; vi) making future scheduling and current data migration selections based upon the hybrid prediction system; and vii) adapting the Bayesian Network Model by feeding the scheduling and data migration selections into said Bayesian Network Model; and D) a memory for storing information related to said occurrence of said critical events and said system parameters of said nodes. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A non-transitory computer readable storage medium comprising program instructions that predicts an occurrence of a critical event in a computer cluster by:
-
implementing a hybrid prediction system comprising rule based prediction algorithms, time-dependent variable prediction algorithms, and a warning window for a computer cluster comprising a series of nodes, said hybrid prediction system carrying out instructions for; maintaining an event log for only those nodes in which an error has occurred, wherein the error adversely affects performance of the cluster or one of its nodes; maintaining a system parameter log for each node in the computer cluster; for only those nodes in which an error has occurred, loading the information from the event log into a Bayesian network model representing a correspondence between the system performance parameters and occurrence of the critical events; for only those nodes in which an error has occurred, loading system performance information pertaining to said error-prone nodes from a system parameter log comprising information concerning system performance parameters for each node in the cluster into the Bayesian network model; using the Bayesian network model to predict a future critical event within a specified time-limit based upon the hybrid prediction system; storing information related to said occurrence of said critical events and said system parameters of said nodes; making future scheduling and current data migration selections based upon the hybrid prediction system; and adapting the Bayesian Network Model by feeding the scheduling and data migration selections into said Bayesian Network Model. - View Dependent Claims (13, 14, 15, 16, 17, 18)
-
Specification