Annotation of network activity through different phases of execution

• US 10,122,605 B2
• Filed: 02/23/2015
• Issued: 11/06/2018
• Est. Priority Date: 07/09/2014
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Kulkarni, Karthik, Nambiar, Raghunath
• Primary Examiner(s): Maskulinski, Michael
• Application Number:
  US14/629,151
• Publication Number:
  US 20160011925A1
• Time in Patent Office: 1,352 Days
• 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 11/3065 Monitoring arrangements determined by the means or processing involved in reporting the monitored data error or fault reporting or logging G06F11/0766
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 43/067 using time frame reporting
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/16 Threshold monitoring

Abstract

The subject technology provides a drillable time-series heat map, which combines information of separate network element (e.g., switch, router, server or storage) and relates them together through impact zones to correlate network wide events and the potential impact on the other units in the network. The subject technology also brings together the network and its components, the distributed application(s) and a heat map controller to proactively communicate with one another to disseminate information such as failures, timeouts, new jobs, etc. For an individual job (e.g., for a distributed application), the subject technology may monitor consumption of resources during different phases of execution to provide individual job profile data that could be presented as a drillable heat map. The heat map, in this regard, nay include resource utilization heat metrics of resources such as CPU, Input/Output (I/O), memory, etc., in the heat map or graphs and presented along with network activity.

Claims

1. A system, comprising:

processor; and

memory including instructions that, when executed by the processor, cause the system to;

receive information for a job to be processed by a distributed application, the job being submitted from a user or other application and having at least two phases of execution for completion of the job;

identify a set of network elements to monitor during processing of the job, the set of network elements corresponding to (a) nodes that are involved in at least a first phase of the job and (b) one or more level of network hierarchy nodes to include potential problems that are created from network higher up in the hierarchy;

monitor, over a period of time, the set of network elements during processing of the job for the at least two phases of execution;

detect a failure during at least one phase of execution in at least one network element;

determine a recursive impact zone of all network elements connected to the at least one network element, in response to detecting the failure;

flag the network elements of the recursive impact zone; and

generate job profile data indicating at least the failure and the network elements of the recursive impact zone;

indicate, in a graphical representation, the failure in the at least one network element based at least in part on the job profile data;

wherein the recursive impact zone includes all further adjoining network elements connected to the at least one network element in a hierarchy to the at least one network element, but does not include adjoining network elements not in the hierarchy to the at least one network element.

2. The system of claim 1, wherein the instructions further cause the processor to:

provide, for display, the graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution.

3. The system of claim 1, wherein to monitor, over the period of time, the set of network elements comprises:

providing a first set of heat maps, wherein each heat map represents actual metrics for the set of network elements during completion of the job; and

providing a set of second heat maps that each second heat map represents estimated or planned metrics for the set of network elements for each of the at least two phases of execution.

4. The system of claim 1, wherein to monitor, over the period of time, the set of network elements further comprises:

determining averages of network, I/O, RAM and CPU utilization of each of the network elements that are monitored during each of the at least two phases of execution.

5. The system of claim 1, wherein to monitor, over the period of time, the set of network elements further comprises:

receiving information related to a phase of execution for the job.

6. The system of claim 5, wherein the information related to the phase of execution comprises at least one of a map phase start, a map phase end, a reduce phase start, or a shuffle phase start.

7. The system of claim 6, wherein the instructions further cause the processor to:

provide, for display, the graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution;

indicate, in a second graphical representation, the information related to the phase of execution; and

provide, for display, the second graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution.

8. The system of claim 1, wherein the set of network elements comprises a set of servers or storage devices in a network hierarchy to be monitored in which data travels through for completing the job.

9. A computer-implemented method, comprising:

receiving information for a job to be processed by a distributed application, the job being submitted from a user or other application and having at least two phases of execution for completion of the job;

identifying a set of network elements to monitor during processing of the job, the set of network elements corresponding to (a) nodes that are involved in at least a first phase of the job and (b) one or more level of network hierarchy nodes to include potential problems that are created from network higher up in the hierarchy;

monitoring, over a period of time, the set of network elements during processing of the job for the at least two phases of execution;

detecting a failure during at least one phase of execution in at least one network element;

determining a recursive impact zone of all network elements connected to the at least one network element; and

generating job profile data indicating at least the failure;

indicating, in a graphical representation, the failure in the at least one network element based at least in part on the job profile data;

wherein the recursive impact zone includes all further adjoining network elements connected to the at least one network element in a hierarchy to the at least one network element, but does not include adjoining network elements not in the hierarchy to the at least one network element.

10. The computer-implemented method of claim 9, further comprising:

providing, for display, the graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution.

11. The computer-implemented method of claim 9, wherein to monitor, over the period of time, the set of network elements further comprises:

collecting monitoring data in an application-network controller, the monitoring data indicating at least one of a utilization of resources and network activity during execution of the job;

providing the monitoring data collected in the application-network controller for inclusion with the job profile data for the job;

providing a first set of heat maps, wherein each heat map represents actual metrics for the set of network elements during completion of the job; and

providing a set of second heat maps that each second heat map represents estimated or planned metrics for the set of network elements for each of the at least two phases of execution.

12. The computer-implemented method of claim 9, wherein to monitor, over the period of time, the set of network elements further comprises:

determining averages of network, I/O, RAM and CPU utilization of each of the network elements that are monitored during each of the at least two phases of execution.

13. The computer-implemented method of claim 9, wherein to monitor, over the period of time, the set of network elements further comprises:

receiving information related to a phase of execution for the job.

14. The computer-implemented method of claim 13, wherein the information related to the phase of execution comprises at least one of a map phase start, a map phase end, a reduce phase start, or a shuffle phase start.

15. The computer-implemented method of claim 14, further comprising:

providing, for display, the graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution;

indicating, in a second graphical representation, the information related to the phase of execution; and

providing, for display, the second graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution.

16. The computer-implemented method of claim 9, wherein the set of network elements comprises a set of servers or storage devices in a network hierarchy to be monitored in which data travels through for completing the job.

17. A non-transitory computer-readable medium including instructions stored therein that, when executed by a computing device, cause the at least one computing device to:

receive information for a job to be processed by a distributed application, the job being submitted from a user or other application and having at least two phases of execution for completion of the job;

identify a set of network elements to monitor during processing of the job, the set of network elements corresponding to (a) nodes that are involved in at least a first phase of the job and (b) one or more level of network hierarchy nodes to include potential problems that are created from network higher up in the hierarchy;

monitor, over a period of time, the set of network elements during processing of the job for the at least two phases of execution;

detect a failure during at least one phase of execution in at least one network element;

determine a recursive impact zone of all network elements connected to the at least one network element, in response to detecting the failure;

flag the network elements of the recursive impact zone; and

generate job profile data indicating at least the failure and the network elements of the recursive impact zone;

indicate, in a graphical representation, the failure in the at least one network element based at least in part on the job profile data;

wherein the recursive impact zone includes all further adjoining network elements connected to the at least one network element in a hierarchy to the at least one network element, but does not include adjoining network elements not in the hierarchy to the at least one network element.

18. The non-transitory computer-readable medium of claim 17, wherein the instructions further cause the at least one computing device to:

provide, for display, the graphical representation indicating the failure at a position along a timeline corresponding to the at least one phase of execution.

19. The non-transitory computer-readable medium of claim 17, wherein to monitor, over the period of time, the set of network elements further comprises:

collecting monitoring data in an application-network controller, the monitoring data indicating at least one of a utilization of resources and network activity during execution of the job;

providing the monitoring data collected in the application-network controller for inclusion with the job profile data for the job;

providing a first set of heat maps, wherein each heat map represents metrics for the set of network elements during completion of the job; and

providing a set of second heat maps that each second heat map represents estimated or planned metrics for the set of network elements for each of the at least two phases of execution.

20. The non-transitory computer-readable medium of claim 17, wherein to monitor, over the period of time, the set of network elements further comprises:

determining averages of network, I/O, RAM and CPU utilization of each of the network elements that are monitored during each of the at least two phases of execution.