Infrastructure performance monitoring
A method and apparatus for infrastructure performance monitoring are described. In one embodiment, the method comprises monitoring virtual application performance, identifying degradation in the application performance, and correlating the degradation to one or more of virtual and physical infrastructure components.
1. A computer-implemented method for infrastructure performance monitoring, the infrastructure comprising physical and virtual components, the method comprising:
monitoring application performance over the infrastructure by communication exchange with the physical and virtual components, including
tapping traffic data using one or more virtual network traffic taps,
duplicating, using the one or more virtual network traffic taps, packets of the traffic data,
forwarding, via a virtual network packet broker, duplicated packets to one or more monitoring tools,
grouping the traffic data based at least on virtual entities in the infrastructure, including correlating the duplicated packets to associate the packets with individual packet flows, and
updating performance counters based on statistics collected from the virtual and physical components using correlated, duplicated packets;
identifying degradation in the application performance due to one or more of the virtual and physical infrastructure components and results of determined correlations between the physical and virtual components of the infrastructure, including directing duplicated packets to one or more analysis tools and using the one or more analysis tools to analyze the grouped traffic data using correlated, duplicated packets to indicate performance according to a virtual infrastructure of which they are part; and
correlating the degradation to the one or more virtual and physical infrastructure components.
2. The method defined in claim 1 wherein correlating the degradation is based on compute, storage, and network statistics.
3. The method defined in claim 1 further comprising:
grouping statistics regarding the virtual and physical infrastructure; and
generating visualization analytics corresponding to grouped statistics.
4. The method defined in claim 3 wherein grouping statistics is according to virtual entity type.
5. The method defined in claim 4 wherein the virtual entity type comprises a port group.
6. The method defined in claim 4 wherein the virtual entity type comprises a data store.
7. The method defined by claim 3 wherein the visualization analytics comprise a user interface that depicts application flows across the infrastructure of physical and virtual components.
8. The method defined by claim 7 wherein the user interface illustrates flow-by-flow through the groups of physical and virtual components of the infrastructure.
9. The method defined in claim 3 further comprising generating a user interface having one or more windows depicting performance statistics with correlations between one or more of:
a first set of one or more physical entities with at least a second set of one or more virtual entities or a third set of one or more physical entities utilized by the physical entities in the first set; and
a fourth set of one or more virtual entities with at least a fifth set of one or more virtual entities or a sixth set of one or more physical entities utilized by the virtual entities in the fourth set.
10. A non-transitory machine-readable storage medium containing executable program instructions which when executed by a data processing system cause the system to monitor an infrastructure comprising physical and virtual components by:
monitoring application performance over the infrastructure by communication exchange with the physical and virtual components, including
tapping traffic data using one or more virtual network traffic taps,
duplicating, using the one or more virtual network traffic taps, packets of the traffic data,
forwarding, via a virtual network packet broker, duplicated packets to one or more monitoring tools,
grouping the traffic data based at least on virtual entities in the infrastructure, including correlating the duplicated packets to associate the packets with individual packet flows, and
updating performance counters based on statistics collected from the virtual and physical components using correlated, duplicated packets;
identifying degradation in the application performance due to one or more of the virtual and physical infrastructure components and results of determined correlations between the physical and virtual components of the infrastructure, including directing duplicated packets to one or more analysis tools and using the one or more analysis tools to analyze the grouped traffic data using correlated, duplicated packets to indicate performance according to a virtual infrastructure of which they are part; and
correlating the degradation to the one or more of virtual and physical infrastructure components.
11. The non-transitory machine-readable storage medium defined in claim 10 wherein correlating the degradation is based on compute, storage, and network statistics.
12. The non-transitory machine-readable storage medium defined in claim 10 further comprising:
grouping statistics regarding the virtual and physical infrastructure; and
generating visualization analytics corresponding to the group statistics.
13. The non-transitory machine-readable storage medium defined in claim 12 wherein grouping statistics is according to virtual entity type.
14. The non-transitory machine-readable storage medium defined in claim 13 wherein the virtual entity type comprises a port group.
15. The non-transitory machine-readable storage medium defined in claim 13 wherein the virtual entity type comprises a data store.
16. The non-transitory machine-readable storage medium defined by claim 12 wherein the visualization analytics comprise a user interface that depicts application flows across the infrastructure of physical and virtual components.
17. The non-transitory machine-readable storage medium defined by claim 16 wherein the user interface illustrates flow-by-flow through the groups of physical and virtual components of the infrastructure.
18. The non-transitory machine-readable storage medium defined by claim 12 wherein the method further comprises generating a user interface having one or more windows depicting performance statistics with correlations between one or more of:
a first set of one or more physical entities with at least a second set of one or more virtual entities or a third set of one or more physical entities utilized by the physical entities in the first set; and
a fourth set of one or more virtual entities with at least a fifth set of one or more virtual entities or a sixth set of one or more physical entities utilized by the virtual entities in the fourth set.