Metric Object Tracking System

• US 20110078227A1
• Filed: 09/28/2009
• Published: 03/31/2011
• Est. Priority Date: 09/28/2009
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: Softlayer Technologies Incorporated (International Business Machines Corporation)
• Inventors: Reese, Joshua Logan, McAloon, Daniel
• Granted Patent: US 9,135,133 B2
• Time in Patent Office: Days
• US Class Current: 709/201
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3093 Configuration details thereof, e.g. installation, enabling, spatial arrangement of the probes
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

Various embodiments of a object metric tracking system are disclosed, applicable to distributed systems composed of a variety of computing nodes coupled together via a network. The computing nodes are arranged in various physical and virtual configurations via a configuration system. Detector nodes are also coupled to the distributed system, each node including one or more modules dedicated to collecting data about tracked system metrics. The data from the detector nodes is sent to a processing engine where it is decomposed for quick storage in a data storage pool including a number of independent storage nodes. A smart metric tracking object is used to provide programmatic access to the data in the storage pool, automatically abstracting away the details of the storage system. A reporter component processes the data and provides human or machine-readable reports for integration with various systems, including a reporting website.

Claims

1. A object metric tracking system comprising:

a distributed system including a plurality of computing nodes coupled to a network;

a configuration system capable of connecting the computing nodes together in a plurality of configurations over the network;

a plurality of detector nodes coupled to the network, each detector node having at least one module associated with a tracked metric and adapted to produce data associated with the tracked metric;

a processing engine coupled to the network and adapted to decompose received metric data into a storable form;

a data storage pool coupled to the network, the data storage pool including at least one independent data storage node adapted to receive and persistently store data associated with at least one tracked metric;

a metric tracking object retrieval system coupled to the network, the metric tracking object retrieval system adapted to retrieve and produce the persistently stored data associated with at least one tracked metric from the data storage pool;

wherein the location of the independent data storage node holding the persistently stored data is automatically calculated by the metric tracking object retrieval system.

2. The metric object tracking system as recited in claim 1, wherein the system further comprises a metric tracking object processor adapted to retrieve the persistently stored data via the metric tracking object retrieval system and format it for reporting.

3. The metric object tracking system as recited in claim 1, wherein the system further comprises a buffer coupled between at least one detector node and the processing engine.

4. The metric object tracking system as recited in claim 1, wherein the computing nodes are organized into at least three separate network configurations by the configuration system based upon the permissions of the users accessing the distributed system.

5. The metric object tracking system as recited in claim 1, wherein a detector includes modules adapted to provide metric data related to one of bandwidth, packet characteristics, storage, node usage, computing time, cache statistics, memory, database use, resources requested, and response time.

6. The metric object tracking system as recited in claim 1, wherein a detector includes modules adapted to provide metric data related to one of temperature, vibration, hard drive failure, voltage, location, equipment utilization, equipment identification, and equipment type.

7. The metric object tracking system as recited in claim 1, wherein a detector includes modules adapted to provide metric data related to one of service cost, quota use, address use, API use, user accounts, email information, available services, services used, billing, DNS information, order status, trouble ticket status, system capacity, network capacity, and network topology.

8. The metric object tracking system as recited in claim 1, wherein a first detector gathers metric data related to a first virtual network and a second detector gathers metric data related to a second virtual network.

9. A metric tracking object retrieval system comprising:

a data storage pool, the data storage pool including at least one independent data storage node adapted to;

receive data associated with at least one tracked metric;

persistently store the data associated with the at least one tracked metric using a computer readable medium; and

read the data stored on the computer readable medium and provide a representation of the data on demand;

a plurality of metric tracking objects, each including;

an API allowing programmatic access to the persistently stored data;

tracking logic operable to automatically calculate the logical location of the persistently stored data in the data storage pool;

a persistence layer including a computer readable medium, and

a persistence interface operable to couple a software representation of a metric tracking object to a representation stored on the computer readable medium via the persistence layer;

in the distributed system including a plurality of computing nodes coupled to a network;

a metric tracking object processor operable to retrieve the persistently stored data via one of the metric tracking objects and format the persistently stored data for reporting;

wherein the persistently stored data associated with the metric tracking objects is retrieved by the metric tracking object processor via a metric tracking object and then stored back in the data storage pool via the metric tracking object.

10. The metric tracking object retrieval system as recited in claim 9, wherein the metric tracking object processor formats a first tracked metric as one of a graph, chart, csv file, html file, and a textual description.

11. The metric tracking object retrieval system as recited in claim 9, wherein the metric tracking object processor is operable to provide updated reports relating to a first tracked metric in real-time or near real-time.

12. The metric tracking object retrieval system as recited in claim 9, wherein the metric tracking object processor further comprises a parallel processing system for correlating metric data retrieved from the data storage pool via a metric tracking object.

13. The metric tracking object retrieval system as recited in claim 9, wherein each metric tracking object further comprises a sharding module and a connection module.

14. The metric tracking object retrieval system as recited in claim 9, wherein the information stored back in the data storage pool via the metric tracking object in transformed.

15. A method for tracking metric data in a distributed system using a metric tracking object, the method comprising:

configuring a distributed system including a plurality of computing nodes into a physical configuration and one or more logical configurations;

associating detectors with a plurality of physical and logical entities in the distributed system;

gathering raw information associated with the physical and logical entities via the detectors;

sending the gathered raw information to a processing engine;

decomposing the gathered raw information into a storable form;

sending the decomposed storable information in a computer readable medium associated with at least one node of a data storage pool;

receiving a request for the tracked metric data via an API;

automatically calculating the location of the decomposed storable information in the data storage pool;

reading decomposed the information from the computer readable medium and providing an electronic copy on demand;

returning information via the API in response to the request.

16. The method as recited in claim 15, wherein the raw information associated with the physical entities is related to one of temperature, vibration, hard drive failure, voltage, location, equipment utilization, equipment identification, equipment type, bandwidth, packet characteristics, storage, node usage, computing time, cache statistics, memory, database use, resources requested, and response time.

17. The method as recited in claim 15, wherein the raw information associated with the logical entities is related to one of service cost, quota use, address use, API use, user accounts, email information, available services, services used, billing, DNS information, order status, trouble ticket status, system capacity, network capacity, network topology, node type, bandwidth, packet characteristics, storage, node usage, computing time, cache statistics, memory, database use, resources requested, and response time.

18. The method as recited in claim 15 further comprising:

transforming the data via a correlation process and saving the information back to the storage pool.

19. The method as recited in claim 15 further comprising:

generating a report from the information.

20. The method as recited in claim 19 wherein the format of the report is one of a graph, chart, csv file, html file, and a textual description.