Back to results
US 12,045,123 B2Granted
Detecting datacenter mass outage with near real-time/offline using ml models
AssigneeOracle International Corporation
InventorsAmarpal Singh Monga, Bin Chen, Alex Edward Hamilton
Life of this patent
Estimating the term…
Abstract
The present embodiments relate to data center outage detection and alert generation. An outage detection service as described herein can process near real-time data from various sources in a datacenter and process the data using a model to determine one or more projected sources of a detected outage. The model as described herein can include one or more machine learning models incorporating a series of rules to process near-real time data and offline data and determine one or more projected sources of an outage. An alert message can be generated to provide the projected sources of the outage and other data relevant to the outage.
Claims19
A method for identifying one or more projected sources of an outage in a datacenter, the method comprising:
- responsive to detecting the outage, processing a set of input data using a model to identify the one or more projected sources of the outage, the model incorporating a set of rules specifying correlations between the set of input data and applications executing on devices as the one or more projected sources of the outage, wherein identifying the one or more projected sources of the outage comprises:
- determining, based at least in part on the set of rules accessible to the model, one or more anomalous parameters within the set of input data;
- and
- identifying one or more potential applications that corresponds to each of a set of determined anomalous parameters, wherein each of the identified one or more potential applications are part of the one or more projected sources of the outage;
- generating an outage notification message including a user interface that comprises a) at least a portion of the identified one or more potential applications, and b) a graph that represents at least a portion of the set of input data and at least a portion of the one or more anomalous parameters;
- identifying, by the model using a first rule of the set of rules, that a first anomalous parameter relates to a first application executing on a portion of servers in the datacenter;
- and
- identifying, by the model using a second rule of the set of rules, that a change to execution of the first application occurred within a threshold time duration of a time of detecting the outage, wherein the first application is included in the outage notification message as a first projected source of the outage, wherein the outage notification message further provides resolution data specifying instructions to revert the first application to a previous version to remove the change to the execution of the first application.
5 more elements
A system configured to identify one or more projected sources of an outage in a datacenter, comprising:
- a processor;
- and
- a non-transitory computer-readable medium including instructions that, when executed by the processor, cause the processor to:
- responsive to detecting the outage, process a set of input data using a model to identify the one or more projected sources of the outage, the model incorporating a set of rules specifying correlations between the set of input data and applications executing on devices as the one or more projected sources of the outage, wherein identifying the one or more projected sources of the outage comprises:
- determining, based at least in part on the set of rules accessible to the model, one or more anomalous parameters within the set of input data;
- and
- identifying one or more potential applications that corresponds to each of a set of determined anomalous parameters, wherein each of the identified one or more potential applications are part of the one or more projected sources of the outage;
- generate an outage notification message including a user interface that comprises a) at least a portion of the identified one or more potential applications, and b) a graph that represents at least a portion of the set of input data and at least a portion of the one or more anomalous parameters;
- identify, by the model using a first rule of the set of rules, that a first anomalous parameter relates to a first application executing on a portion of servers in the datacenter;
- and
- identify, by the model using a second rule of the set of rules, that a change to execution of the first application occurred within a threshold time duration of a time of detecting the outage, wherein the first application is included in the outage notification message as a first projected source of the outage, wherein the outage notification message further provides resolution data specifying instructions to revert the first application to a previous version to remove the change to the execution of the first application.
8 more elements
One or more non-transitory computer-readable medium including stored thereon a sequence of instructions that, when executed by one or more processors configured to identify one or more projected sources of an outage in a datacenter, causes the one or more processors to perform operations comprising:
- responsive to detecting the outage, processing a set of input data using a model to identify the one or more projected sources of the outage, the model incorporating a set of rules specifying correlations between the set of input data and applications executing on devices as the one or more projected sources of the outage, wherein identifying the one or more projected sources of the outage comprises:
- determining, based at least in part on the set of rules accessible to the model, one or more anomalous parameters within the set of input data;
- and
- identifying one or more potential applications that corresponds to each of a set of determined anomalous parameters, wherein each of the identified one or more potential applications are part of the one or more projected sources of the outage;
- identifying, by the model using a first rule of the set of rules, that a first anomalous parameter relates to a first application executing on a portion of servers in the datacenter;
- identifying, by the model using a second rule of the set of rules, that a change to execution of the first application occurred within a threshold time duration of a time of detecting the outage, wherein the first application is included in the outage notification message as a first projected source of the outage, wherein the outage notification message further provides resolution data specifying instructions to revert the first application to a previous version to remove the change to the execution of the first application;
- and
- generating an outage notification message including a user interface that comprises a) at least a portion of the identified one or more potential applications, and b) a graph that represents at least a portion of the set of input data and at least a portion of the one or more anomalous parameters.
5 more elements
Dependencies are inferred from the claim text.