Oversubscription monitor
View Patent ↗Aspects of oversubscription monitoring are described. In one embodiment, oversubscription monitoring includes accumulating an amount of data that arrives at a network component over at least one epoch of time. Further, a core processing rate at which data can be processed by the network component is calculated. Based on the amount of data and the core processing rate, it is determined whether the network component is operating in an oversubscribed region of operation. In one embodiment, when the network component is operating in the oversubscribed region of operation, certain quality of service metrics are monitored. Using the monitored metrics, a network operation display object may be generated for identifying or troubleshooting network errors during an oversubscribed region of operation of the network component.
1. A method of oversubscription monitoring, comprising:
accumulating an amount of data that arrives at a network component over an epoch of time;
calculating a core processing rate at which the data can be processed by the network component;
determining whether the network component is operating in an oversubscribed region of operation based on the amount of data and the core processing rate;
when the network component is operating in the oversubscribed region of operation, measuring at least one quality of service metric of the network component; and
generating a network operation display object that identifies whether the network component operated in the oversubscribed region over at least one epoch of time.
2. The method of claim 1 , wherein measuring at least one quality of service metric comprises counting a number of data drop events and a number of flow control events of the network component.
3. The method of claim 2 , wherein counting a number of data drop events comprises counting a number of byte or packet data drop events per port of the network component.
4. The method of claim 3 , further comprising determining a maximum usage of an oversubscription buffer of the network component over the epoch of time.
5. The method of claim 4 , wherein generating a network operation display object further comprises generating an object that identifies, for each of a plurality of epochs of time over a time window, the amount of data that arrives at the network component, the number of data drop events, the number of flow control events, and the maximum usage of the oversubscription buffer of the network component.
6. The method of claim 4 , further comprising generating a consolidated data batch by consolidating, over each of a plurality of epochs of time, the amount of data that arrives at the network component, the number of data drop events, the number of flow control events, and the maximum usage of the oversubscription buffer of the network component.
7. The method of claim 6 , further comprising transmitting the consolidated data batch to a data aggregator that collects oversubscription data from a plurality of network components.
8. The method of claim 1 , wherein the accumulating comprises accumulating, for each of a plurality of epochs of a sampling window, the amount of data that arrives at the network component over the epoch of time.
9. The method of claim 1 , wherein:
accumulating an amount of data comprises accumulating, for each port of a pipeline of the network component, an amount of data that arrives at the port over the epoch of time; and
the method further comprises aggregating the amount of data accumulated for each port in the pipeline over the epoch of time, to determine an amount of data that arrives at the pipeline of the network component over the epoch of time.
10. A system for oversubscription monitoring, comprising:
an oversubscription buffer of a network component that buffers received data for processing; and
an oversubscription monitor that:
accumulates an amount of data that arrives at the network component over an epoch of time;
determines whether the network component is operating in an oversubscribed region of operation based on a core processing rate of the network component;
measures at least one quality of service metric of the network component; and
generates a network operation display object that includes a representation of the at least one quality of service metric of the network component and identifies whether the network component operated in the oversubscribed region over the epoch of time.
11. The system of claim 10 , wherein the oversubscription monitor further counts a number of data drop events and a number of flow control events of the network component.
12. The system of claim 11 , wherein the oversubscription monitor further counts a number of byte or packet data drop events per port of the network component.
13. The system of claim 10 , wherein the oversubscription monitor further determines a maximum usage of the oversubscription buffer over the epoch of time.
14. The system of claim 10 , wherein the oversubscription monitor further accumulates, for each of a plurality of epochs of a sampling window, the amount of data that arrives at the network component over the epoch of time.
15. The system of claim 10 , wherein the oversubscription monitor further:
accumulates, for each port of a pipeline of the network component, an amount of data that arrives at the port over the epoch of time; and
aggregates the amount of data accumulated for each port in the pipeline over the epoch of time, to determine an amount of data that arrives at the pipeline of the network component over the epoch of time.
16. A method of oversubscription monitoring, comprising:
accumulating an amount of data that arrives at a network component and calculating a core processing rate at which data can be processed by the network component over an epoch of time;
when the network component is operating in an oversubscribed region of operation, measuring at least one quality of service metric of the network component; and
generating a network operation display object that identifies whether the network component operated in the oversubscribed region over at least one epoch of time.
17. The method of claim 16 , wherein measuring at least one quality of service metric comprises:
counting a number of data drop events and a number of flow control events of the network component; and
marking a packet of data that arrives at the network component to indicate congestion at the network component based on the number of data drop events or the number of flow control events.
18. The method of claim 17 , wherein counting the number of data drop events comprises counting a number of byte or packet data drop events per port of the network component.
19. The method of claim 16 , further comprising determining a maximum usage of an oversubscription buffer of the network component over the epoch of time.
20. The method of claim 18 , wherein generating a network operation display object further comprises generating an object that identifies, for each of a plurality of epochs of time over a time window, the amount of data that arrives at the network component, the number of data drop events, the number of flow control events, and the maximum usage of the oversubscription buffer of the network component.