Information technology resource management
Embodiments relate to information technology resource management and scaling. According to one aspect, an upcoming event impacting an application operating on one or more resources is identified. A workload on the application is predicted based on the upcoming event identified and historical data about a previous event having similarity with the upcoming event. The prediction is performed using a predefined rule. A number of resources required to process the predicted workload is ascertained using a past scaling history of the application. The resources are scaled based on the ascertained number of said resources determined before the occurrence of the event.
1. A system comprising:
a memory having computer readable instructions; and
a processor for executing the computer readable instruction, the instructions including:
identifying an upcoming event impacting an application operating on one or more resources, the identifying including retrieving information about the upcoming event from a database;
predicting a workload on the application based on the upcoming event identified, a popularity score of the upcoming event, and a historical data of a previous event having similarity with the upcoming event, wherein the prediction is performed using a predefined rule;
comparing the predicted workload with an actual historical workload on the application;
verifying the prediction based on the comparison;
ascertaining a number of said resources required to process the predicted workload using a past scaling history of the application;
determining a time duration for scaling said resources to the number of said resources ascertained; and
scaling said resources based on the ascertained number of said resources before the occurrence of the event, wherein the scaling said resources is initialized at a time that proceeds the occurrence of the event by at least the time duration.
2. The system of claim 1 , wherein the instructions further include:
monitoring one or more information streams received from one or more sources for identifying the upcoming event.
3. The system of claim 1 , wherein the instructions further include:
computing the popularity score of the upcoming event based on contents of a social media stream.
4. The system of claim 3 , wherein the information streams include the social media stream.
5. The system of claim 1 , wherein the historical data comprises a historical workload on the application corresponding to the previous event.
6. The system of claim 1 , wherein the historical data comprises historical information corresponding to the previous event.
7. The system of claim 1 , wherein the information streams corresponds to a web based information, a sensor data, an audio stream, a video stream, or a radio audio stream.
8. The system of claim 1 , wherein:
the information about the upcoming event is generated based on information streams from one or more sources that are processed by a stream processing engine; and
the information about the upcoming event is stored in the database by the stream processing engine.
9. The system of claim 8 , wherein the information streams include at least one stream having information selected from the group consisting of web based information, sensor data, an audio stream, a video stream, and a radio audio stream.
10. A system comprising:
a memory having computer readable instructions; and
a processor for executing the computer readable instruction, the instructions including:
identifying an upcoming event impacting an application operating on one or more resources, the identifying based on an information source that is external to the system;
predicting a workload on the application based on the upcoming event identified, a popularity score of the upcoming event, and a historical data of a previous event having similarity with the upcoming event, wherein the prediction is performed using a predefined rule;
comparing the predicted workload with an actual historical workload on the application;
verifying the prediction based on the comparison;
ascertaining a number of said resources required to process the predicted workload using a past scaling history of the application;
determining a time duration for scaling said resources to the number of said resources ascertained; and
scaling said resources based on the ascertained number of said resources before the occurrence of the event, wherein the scaling said resources is initialized at a time that proceeds the occurrence of the event by at least the time duration.