Congestion avoidance in a slice-based network
A system can reduce congestion in slice-based networks, such as a virtual service network (“VSN”). The system can include a monitoring module that communicates with agents on switches, such as routers or servers. The switches report telematics data to the monitoring module, which determines slice-specific performance attributes such as slice latency and slice throughput. These slice-specific performance attributes are compared against software license agreement (“SLA”) requirements. When the SLA is not met, the monitoring module can implement a new slice path for the slice to reduce the congestion.
1. A method for managing latency in a slice-based network, comprising:
setting a flag in a packet header for a packet that corresponds to a slice;
receiving slice-specific packet timing information from multiple switches in a slice-based network, the packet timing information being determined by agents executing on the switches, wherein the agents determine the slice-specific packet timing information by at least:
when the flag is recognized but the packet does not correlate to a flow already stored at the switch, storing the flow in association with a current time;
when the flow is recognized as stored at the respective switch based on address information in the packet header, deducting the stored current time from a new current time; and
removing the flow from storage at the switch;
determining a latency value based on the packet timing information from switches in a first slice path for the slice;
based on comparing the latency value to a threshold, selecting an alternate slice path for the slice, the alternate path including a third switch that is not in the first slice path; and
implementing the alternate slice path across at least the first and third switches.
2. The method of claim 1 , wherein the packet timing information is based on the agents reporting timing for hops at the switches, wherein the packet timing information is received by a monitoring module and added together as part of determining the latency value for the slice.
3. The method of claim 1 , wherein the threshold is established by a service level agreement (“SLA”) that includes a maximum latency for the slice.
4. The method of claim 1 , wherein the switches report the packet timing information based on recognizing the flag in a transmission control protocol (“TCP”) packet header.
5. The method of claim 1 , wherein receiving packet timing information includes receiving source and destination addresses from the first and second agents, wherein a monitoring module uses the source and destination addresses as part of calculating the latency value for the slice.
6. The method of claim 1 , wherein calculating the latency value includes averaging latency for the slice over a period of time.
7. The method of claim 1 , wherein implementing the alternate slice path includes instantiating a new virtual network function (“VNF”) for access by slice traffic at the third switch prior to setting the first switch to route slice traffic to the third switch.
8. A non-transitory, computer-readable medium comprising
instructions that, when executed by a processor, perform stages for managing latency in a slice-based network, the stages comprising:
setting a flag in a packet header for a packet that corresponds to a slice;
receiving slice-specific packet timing information from multiple switches in a slice-based network, the packet timing information being determined by agents executing on the switches, wherein the agents determine the slice-specific packet timing information by at least:
when the flag is recognized but the packet does not correlate to a flow already stored at the switch, storing the flow in association with a current time;
when the flow is recognized as stored at the respective switch based on address information in the packet header, deducting the stored current time from a new current time; and
removing the flow from storage at the switch;
determining a latency value based on the packet timing information from switches in a first slice path for the slice;
based on comparing the latency value to a threshold, selecting an alternate slice path for the slice, the alternate path including a third switch that is not in the first slice path; and
implementing the alternate slice path across at least the first and third switches.
9. The non-transitory, computer-readable medium of claim 8 , wherein the packet timing information is based on the agents reporting timing for hops at the switches, wherein the packet timing information is received by a monitoring module and added together as part of determining the latency value for the slice.
10. The non-transitory, computer-readable medium of claim 8 , wherein the threshold is established by a service level agreement (“SLA”) that includes a maximum latency for the slice.
11. The non-transitory, computer-readable medium of claim 8 , wherein the switches report the packet timing information based on recognizing the flag in a transmission control protocol (“TCP”) packet header.
12. The non-transitory, computer-readable medium of claim 8 , wherein receiving packet timing information includes receiving source and destination addresses from the first and second agents, wherein a monitoring module uses the source and destination addresses as part of calculating the latency value for the slice.
13. The non-transitory, computer-readable medium of claim 8 , wherein calculating the latency value includes averaging latency for the slice over a period of time.
14. The non-transitory, computer-readable medium of claim 8 , wherein implementing the alternate slice path includes instantiating a new virtual network function (“VNF”) for access by slice traffic at the third switch prior to setting the first switch to route slice traffic to the third switch.
15. A system for managing in a slice-based network, comprising:
a non-transitory, computer-readable medium containing instructions for a monitoring module; and
a processor that executes the monitoring module to perform stages comprising:
setting a flag in a packet header for a packet that corresponds to a slice;
receiving slice-specific packet timing information from multiple switches in a slice-based network, the packet timing information being determined by agents executing on the switches, wherein the agents determine the slice-specific packet timing information by at least:
when the flag is recognized but the packet does not correlate to a flow already stored at the switch, storing the flow in association with a current time;
when the flow is recognized as stored at the respective switch based on address information in the packet header, deducting the stored current time from a new current time; and
removing the flow from storage at the switch;
determining a latency value based on the packet timing information from switches in a first slice path for the slice;
based on comparing the latency value to a threshold, selecting an alternate slice path for the slice, the alternate path including a third switch that is not in the first slice path; and
implementing the alternate slice path across at least the first and third switches.
16. The system of claim 15 , wherein the packet timing information is based on the agents reporting timing for hops at the switches, wherein the packet timing information is received by a monitoring module and added together as part of determining the latency value for the slice.
17. The system of claim 15 , wherein the threshold is established by a service level agreement (“SLA”) that includes a maximum latency for the slice.
18. The system of claim 15 , wherein the switches report the packet timing information based on recognizing the flag in a transmission control protocol (“TCP”) packet header.
19. The system of claim 15 , wherein receiving packet timing information includes receiving source and destination addresses from the first and second agents, wherein a monitoring module uses the source and destination addresses as part of calculating the latency value for the slice.
20. The system of claim 15 , wherein calculating the latency value includes averaging latency for the slice over a period of time.