SYSTEMS, METHODS, AND COMPUTER PROGRAMS PRODUCTS PROVIDING RELEVANT CORRELATION OF DATA SOURCE PERFORMANCE

• US 20150178176A1
• Filed: 03/05/2014
• Published: 06/25/2015
• Est. Priority Date: 12/20/2013
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Joshi, Aashay, Velella, Alberto, Sprague, Jason, Gilbao, Ran, Gabay, Yarom
• Granted Patent: US 9,367,421 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• 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 11/3051 Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs verification or detection of system hardware configuration G06F11/2247
• CPC Class Codes: G06F 11/323 Visualisation of programs or trace data
• CPC Class Codes: G06F 11/328 Computer systems status display G06F11/327 takes precedence
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 2201/865 Monitoring of software
• CPC Class Codes: G06F 8/60 Software deployment
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57
• CPC Class Codes: G06F 8/71 Version control security arrangements therefor G06F21/57; Configuration management
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/22 comprising specially adapted graphical user interfaces [GUI]
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/04 Processing captured monitoring data, e.g. for logfile generation
• CPC Class Codes: H04L 43/045 for graphical visualisation of monitoring data
• CPC Class Codes: H04L 43/065 related to network devices
• CPC Class Codes: H04L 43/0894 : Packet rate
• CPC Class Codes: H04L 67/34 : involving the movement of software or configuration parameters network booting or remote initial program loading [RIPL] G06F9/4416
• CPC Class Codes: H04L 67/75 : Indicating network or usage conditions on the user display

Abstract

A method performed by a monitoring tool in a computer system, the method including: for a set of network nodes in a computer system: applying a correlation formula on an input based on performance data of the set, and determining a correlation score based on applying the correlation formula, the correlation score indicating a correlation between network nodes in the set; determining, based on the correlation scores, a first list including a first plurality of network nodes having a correlation score that satisfies a first threshold; identifying a second plurality of network nodes included in the first list, the second plurality of network nodes having a correlation score that satisfies a second threshold, which indicates a correlation that is higher than the first threshold; analyzing the performance data of the second plurality against a constancy metric; and removing, based on the analyzing, the second plurality from the first list.

Claims

1. A method of determining a correlation between network nodes in a computer system, comprising:

for a set of network nodes in a computer system;

applying a correlation formula on an input based on performance data of the set of network nodes; and

determining a correlation score based on the applying a correlation formula, the correlation score indicating a correlation between network nodes in the set of network nodes;

determining, based on the correlation scores, a first list including a first plurality of network nodes having a correlation score that satisfies a first threshold;

identifying a second plurality of network nodes included in the first list, the second plurality of network nodes having a correlation score that satisfies a second threshold, the second threshold indicating a correlation that is higher than the first threshold;

analyzing the performance data of the second plurality of network nodes against a constancy metric; and

removing, based on the analyzing, the second plurality of network nodes from the first list, thereby creating a second list.

2. The method of claim 1, further including:

determining a third list including correlation scores between network nodes in the set of network nodes; and

identifying in the third list a third plurality of network nodes having a correlation score that does not satisfy the first threshold; and

removing the third plurality of network nodes from the third list, thereby creating the first list.

3. The method of claim 1, wherein each network node comprises at least one of:

a virtual volume, storage drive, storage controller, server, host machine, virtual machine, or an application.

4. The method of claim 1, further comprising generating the second of network nodes by including topologically-related nodes and excluding nodes that are not topologically related.

5. The method of claim 1, further including:

obtaining network topology data on the set of network nodes; and

receiving the performance data based on one or more operations at the set of network nodes.

6. The method of claim 5, wherein receiving the performance data includes receiving raw data based on one or more operations at a first network node of the set of network nodes, the method further including:

storing the raw data; and

generating summary data for the first network node, the summary data being derived from the raw data, and the input including the summary data.

7. The method of claim 1, wherein for the set of network nodes including a first network node and a second network node, the applying a correlation formula determines a performance correlation between the first network node and the second network node.

8. The method of claim 7, wherein the first network node is of a first type and the second network node is of a second type, wherein the applying a correlation formula includes applying a first performance metric for the first network node of the first type and applying a second performance metric for the second network node of the second type.

9. The method of claim 8, wherein the first performance metric comprises at least one of:

a central processing unit (CPU) counter, I/O operations counter, port traffic, average I/O operations, or maximum I/O operations.

10. The method of claim 7, wherein the first network node is from a different vendor than the second network node.

11. The method of claim 1, further including:

selecting N correlation scores included in the second list, the selected N correlation scores indicating the highest N correlations in the second list; and

for each of the N correlation scores;

identifying a third plurality of network nodes having the respective correlation score; and

displaying the third plurality of network nodes and the respective correlation score.

12. The method of claim 1, wherein the analyzing the performance data includes:

removing the second plurality of network nodes from the first list in response to determining that constancy values of each of the network nodes of the second plurality exceed a constancy threshold.

13. The method of claim 1, wherein a first network node of the set of network nodes maintains a native performance counter, and a second network node of the set of network nodes does not maintain a native performance counter.

14. A system for determining a correlation between network nodes in a computer system, comprising:

a correlation module that for a plurality of topologically related network nodes in a computer system, applies a correlation formula to performance data of the respective network nodes to calculate correlation scores for each of the network nodes, wherein each of the correlation scores indicates a correlation of performance data between two network nodes of the plurality of topologically related network nodes;

a first filter module that determines, based on the one or more correlation scores, a first list including a first subset of network nodes having a correlation score that satisfies a first threshold; and

a second filter module that identifies a second subset of network nodes included in the first subset and determines, based on a constancy metric of the performance data, whether to remove ones of the nodes within the second subset.

15. The system of claim 14, wherein the network topology data includes communication data on which network nodes communicate with ones of the network nodes, and wherein network nodes that communicate with each other are determined to be related to each other.

16. The system of claim 14, wherein the input is based on network topology data, the method further including:

a network topology extraction module that identifies, based on the network topology data, topologically-related network nodes.

17. The system of claim 14, wherein the correlation module determines a correlation between a first network node and a second network node of the set, and wherein the second network node does not maintain a native performance counter.

18. The system of claim 17, wherein the correlation module identifies a second plurality of network nodes that is closely related to the second network node, and wherein each network node of the second plurality of network nodes maintains a native performance counter.

19. The system of claim 18, wherein for each network node of the second plurality of network nodes, the correlation module determines a correlation score between the respective network node and the first network node and uses each of the correlation scores to calculate a correlation score for the application.

20. A computer program product having a computer readable medium tangibly recording computer program logic for analyzing performance data of a computer system, the computer program product comprising:

code to, for each network node within a plurality of topologically-related network nodes in a computer system;

apply a correlation formula on performance data of a respective network node; and

determine a correlation score from the correlation formula, the correlation score indicating a correlation between the respective network node and another network node within the in the plurality of topologically-related network nodes;

code to determine, based on the correlation scores, a first list including a subset of network nodes having a correlation score that satisfies a first threshold;

code to identify a second subset of network nodes included in the first list and having a correlation score that satisfies a second threshold, the second threshold indicating a correlation that is higher than the first threshold;

code to determine, based on the performance data of the plurality of network nodes, whether to remove the second subset of network nodes from the first list; and

code to remove, based on the determining whether to remove, the second subset of network nodes from the first list, thereby generating a second list.