Datacenter operations using search and analytics

• US 9,767,197 B1
• Filed: 08/20/2014
• Issued: 09/19/2017
• Est. Priority Date: 08/20/2014
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Agarwal, Shiv, Gupta, Apurv, Gupta, Mukul, Sharma, Abhijit, Toshniwal, Rohit
• Primary Examiner(s): LEROUX, ETIENNE PIERRE
• Application Number:
  US14/464,579
• Time in Patent Office: 1,126 Days
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A datacenter management system uses data collection proxies to collect performance data and configuration data for different physical and virtual entities in the datacenter. A schema is used to represent the different entities, entity relationships, and entity properties in the datacenter. A search engine identifies the intent of a natural language based search query based on the schema and a datacenter dictionary. The search engine then searches the data based on the search query intent. A dictionary manager converts both periodic and aperiodic data into a time series. This allows the search engine to operate as a time machine identifying both performance data and configuration data for any selectable time period.

Claims

1. A method, comprising:

receiving, by the computing device, a natural language based search query including a phrase or an abbreviation for information associated with managing operation of a datacenter;

identifying, by the computing device, physical connections and hierarchal relationships of physical entities of the datacenter relevant to the natural language based search query;

identifying, by the computing device, hierarchal relationships of virtual entities of the datacenter relevant to the natural language based search query;

identifying, by the computing device, an intent of the natural language based search query based on the phrase or the abbreviation and the identified physical connections and hierarchal relationships of the physical entities and the identified hierarchal relationships of the virtual entities;

performing a natural language based search, by the computing device, of data pertaining to operation of the data center based on the intent of the natural language based search query; and

generating results of the search other than results from a keyword based search.

2. The method of claim 1, further comprising storing, by the computing device, configuration data and performance data associated with the physical entities and the virtual entities.

3. The method of claim 2, further comprising:

converting the configuration data and the performance data associated with aperiodic events into a time series; and

combining the configuration data and the performance data associated with the aperiodic events with the configuration data and the performance data associated with periodic events.

4. The method of claim 3, wherein converting the data associated with aperiodic events into a time series comprises:

generating a first document version for the configuration data and the performance data; and

generating a second document version for the configuration data and the performance data in response to one of the aperiodic events.

5. The method of claim 1, further comprising:

providing, by the computing device, a datacenter dictionary containing keywords, sentence structures, and grammar associated with the datacenter, wherein the grammar associates an abbreviation with a part of speech including a noun or a verb; and

identifying, by the computing device, the intent of the natural language based search query based on the datacenter dictionary.

6. The method of claim 1, further comprising:

identifying, by the computing device, a network configured in the datacenter; and

visualizing, by the computing device, a topology for the network.

7. The method of claim 6, wherein visualizing the topology includes displaying a series of layers representing at least one of the physical entities and the virtual entities associated with the network.

8. The method of claim 1, wherein the results of the search include a state of a network, and

wherein the method further comprises:

storing, by the computing device, the state of the network in a collaboration database; and

enabling, by the computing device, access and manipulation of the state of the network by different users.

9. The method of claim 1, further comprising:

identifying, by the computing device, a network associated with the datacenter;

identifying, by the computing device, two of the physical entities associated with the network;

identifying, by the computing device, ports associated with the two of the physical entities associated with the network; and

identifying, by the computing device, mismatches between the ports.

10. An apparatus, comprising:

a processing device configured to;

associate configuration data and performance data for physical entities and virtual entities of a datacenter;

receive a natural language based search query including a phrase or an abbreviation for information associated with managing operation of the datacenter;

identify physical connections and hierarchal relationships of physical entities of the datacenter relevant to the natural language based search query;

identify hierarchal relationships of virtual entities of the datacenter relevant to the natural language based search query;

identify an intent of the natural language based search query based on the phrase or the abbreviation and the identified physical connections and hierarchal relationships of the physical entities and the identified hierarchal relationships of the virtual entities;

generate an execution plan for searching the configuration data and the performance data based on the based on the intent of the natural language based search query; and

generate results based on the execution plan other than results from a keyword based search.

11. The apparatus of claim 10, wherein the processing device is further configured to:

periodically extract at least some of the performance data for the physical entities and the virtual entities; and

aperiodically extract changes in the configuration data from the physical entities and the virtual entities.

12. The apparatus of claim 11, wherein the processing device is further configured to convert the periodically extracted performance data and aperiodic changes in the configuration data into time-series data.

13. The apparatus of claim 10, wherein the processing device is further configured to:

generate a first document version for the configuration data and performance data associated with a first one of the physical entities or the virtual entities;

detect an event associated with the performance data and/or configuration data for the first one of the physical entities or the virtual entities;

associate a first time period with the first document version ending at a time associated with the event;

generate a second document version for the configuration data and performance data associated with the first one of the physical entities or the virtual entities; and

associate a second time period with the second document version starting at the time associated with the event.

14. The apparatus of claim 13, wherein the processing device is further configured to:

identify a request time associated with the natural language based search query;

compare request time with the first time period for the first document version;

identify in the results the performance data and/or configuration data in the first document version based on an overlap of the request time with the first time period;

compare the request time with the second time period for the second document version; and

identify in the results the performance data and/or configuration data in the second document version based on an overlap of the search time with the second time period.

15. The apparatus of claim 10, wherein the processing device is further configured to visualize the results based on the execution plan.

16. A system, comprising:

proxies configured to collect data for physical entities and virtual entities associated with a datacenter, wherein the data includes configuration data and performance data;

a dictionary comprising keywords, sentence structures, and grammar associated with the datacenter, wherein the grammar associates an abbreviation with a part of speech including a noun or a verb; and

a search engine configured to;

receive a natural language based search query including a phrase or an abbreviation for information associated with managing operation of the datacenter;

identify physical connections and hierarchal relationships of physical entities of the datacenter relevant to the natural language based search query;

identify hierarchal relationships of virtual entities of the datacenter relevant to the natural language based search query;

identify an intent of the natural language based search query based on the phrase or the abbreviation and the identified physical connections and hierarchal relationships of the physical entities and the identified hierarchal relationships of the virtual entities;

generate a search plan based on the natural language based search query, the phrase or the abbreviation, and the dictionary; and

generate results based on the search plan other than results from a keyword based search.

17. The system of claim 16, further comprising a model manager configured to convert periodic performance data for the physical entities and virtual entities and aperiodic configuration data for the physical entities and virtual entities into time-series data.

18. The system of claim 17, wherein the search engine is configured to identify a time period associated with the natural language based search query and identify the time-series data associated with the time period.

19. The system of claim 16, further comprising a visualization manager configured to visualize the results based on the search plan.

20. The system of claim 16, further comprising an analytics engine configured to identify anomalies and problems associated with the physical entities and the virtual entities based on the performance data, the configuration data, and/or models associated with the physical entities and the virtual entities.

21. The system of claim 20, further comprising a visualization manager configured to display the identified anomalies and problems within a topology showing a corresponding physical entity or virtual entity located in the datacenter.

22. An apparatus, comprising:

a processing device configured to;

use a schema to identify relationships and properties for entities in a datacenter;

use a data model to associate configuration data and performance data for the entities in the datacenter;

receive a search query;

generate an execution plan for searching the configuration data and performance data based on the schema and the data model;

generate a first document version for the configuration data and performance data associated with a first one of the entities;

detect an event associated with the performance data and/or configuration data for the first one of the entities;

associate a first time period with the first document version ending at a time associated with the event;

generate a second document version for the configuration data and performance data associated with the first one of the entities;

associate a second time period with the second document version starting at the time associated with the event;

identify a request time associated with the search query;

compare request time with the first time period for the first document version;

identify in search results the performance data and/or configuration data in the first document version based on an overlap of the request time with the first time period;

compare the request time with the second time period for the second document version; and

identify in the search results the performance data and/or configuration data in the second document version based on an overlap of the search time with the second time period.