Dynamic slice bandwidth multiplexing based on slice priority
In a slice-based network, slice multiplexers can be used to anchor inter-cloud tunnels across different clouds in a slice path. The slice multiplexers can dynamically change a total allocated bandwidth of an outer tunnel and reconfigure relative slice bandwidths of inner tunnels. This can result in an optimized bandwidth allocation that enforces slice priorities, maintains required SLA performance levels, and minimizes total allocated bandwidth on the network connection. The dynamic changes can be based on slice priority levels, total number of slices, and historical slice throughput.
1. A method for slice-based quality of service (“QoS”), comprising:
creating a network connection between first and second slice multiplexers at different clouds;
determining that a first slice is utilizing less bandwidth across the network connection than allotted by a service level agreement (“SLA”);
based on the determination, reconfiguring a first slice multiplexer to increase an allowed bandwidth of a second slice and decrease an allowed bandwidth of the first slice, wherein the increase and decrease is based on relative slice priority levels of the first and second slices;
comparing a current bandwidth of the network connection to a total allowed bandwidth, wherein the current bandwidth accounts for the first and second slices; and
based on the comparison, changing the total allowed bandwidth of the network connection by setting limits at the first and second slice multiplexers, wherein the slice priority levels of the first and second slices is factored into determining the changed total allowed bandwidth.
2. The method of claim 1 , further comprising setting a crossbar switch to route packets between multiple slice multiplexers including the first slice multiplexer, wherein the routing depends on a slice path for the packets.
3. The method of claim 1 , wherein the network connection is an intercloud tunnel having an outer tunnel and multiple inner tunnels, the inner tunnels corresponding to different slices,
wherein the change to the total allowed bandwidth applies to the outer tunnel, and
wherein the decrease and increase of bandwidth for the first and second slices, respectively, applies to at least some of the inner tunnels.
4. The method of claim 1 , wherein the network connection includes a path through an MPLS network, and wherein links are changed to reroute the network connection based on which links reach bandwidth capacity.
5. The method of claim 1 , wherein the changed total allowed bandwidth is also based on historical slice throughput.
6. The method of claim 1 , further comprising mirroring the reconfiguration and changed total bandwidth of the first slice multiplexer at the second slice multiplexer.
7. The method of claim 1 , wherein the changed total allowed bandwidth is further based on a number of slices passing through the network connection.
8. A non-transitory, computer-readable medium comprising instructions that are executed by a processor to perform stages for slice-based quality of service (“QoS”), the stages comprising:
creating a network connection between first and second slice multiplexers at different clouds;
determining that a first slice is utilizing less bandwidth across the network connection than allotted by a service level agreement (“SLA”);
based on the determination, reconfiguring a first slice multiplexer to increase an allowed bandwidth of a second slice and decrease an allowed bandwidth of the first slice, wherein the increase and decrease is based on relative slice priority levels of the first and second slices;
comparing a current bandwidth of the network connection to a total allowed bandwidth, wherein the current bandwidth accounts for the first and second slices; and
based on the comparison, changing the total allowed bandwidth of the network connection by setting limits at the first and second slice multiplexers, wherein the slice priority levels of the first and second slices is factored into determining the changed total allowed bandwidth.
9. The non-transitory, computer-readable medium of claim 8 , the stages further comprising setting a crossbar switch to route packets between multiple slice multiplexers including the first slice multiplexer, wherein the routing depends on a slice path for the packets.
10. The non-transitory, computer-readable medium of claim 8 , wherein the network connection is an intercloud tunnel having an outer tunnel and multiple inner tunnels, the inner tunnels corresponding to different slices,
wherein the change to the total allowed bandwidth applies to the outer tunnel, and
wherein the decrease and increase of bandwidth for the first and second slices, respectively, applies to at least some of the inner tunnels.
11. The non-transitory, computer-readable medium of claim 8 , wherein the network connection includes a path through an MPLS network, and wherein links are changed to reroute the network connection based on which links reach bandwidth capacity.
12. The non-transitory, computer-readable medium of claim 8 , wherein the changed total allowed bandwidth is also based on historical slice throughput.
13. The non-transitory, computer-readable medium of claim 8 , the stages further comprising mirroring the reconfiguration and changed total bandwidth of the first multiplexer the second slice multiplexer.
14. The non-transitory, computer-readable medium of claim 8 , wherein the changed total allowed bandwidth is further based on a number of slices passing through the network connection.
15. A system for slice-based quality of service (“QoS”), comprising:
a non-transitory, computer-readable medium containing instructions; and
a processor that executes the instructions to perform stages comprising:
creating a network connection between first and second slice multiplexers at different clouds;
determining that a first slice is utilizing less bandwidth across the network connection than allotted by a service level agreement (“SLA”);
based on the determination, reconfiguring a first slice multiplexer to increase an allowed bandwidth of a second slice and decrease an allowed bandwidth of the first slice, wherein the increase and decrease is based on relative slice priority levels of the first and second slices;
comparing a current bandwidth of the network connection to a total allowed bandwidth, wherein the current bandwidth accounts for the first and second slices; and
based on the comparison, changing the total allowed bandwidth of the network connection by setting limits at the first and second slice multiplexers, wherein the slice priority levels of the first and second slices is factored into determining the changed total allowed bandwidth.
16. The system of claim 15 , the stages further comprising setting a crossbar switch to route packets between multiple slice multiplexers including the first slice multiplexer, wherein the routing depends on a slice path for the packets.
17. The system of claim 15 , wherein the network connection is an intercloud tunnel having an outer tunnel and multiple inner tunnels, the inner tunnels corresponding to different slices,
wherein the change to the total allowed bandwidth applies to the outer tunnel, and
wherein the decrease and increase of bandwidth for the first and second slices, respectively, applies to at least some of the inner tunnels.
18. The system of claim 15 , wherein the network connection includes a path through an MPLS network, and wherein links are changed to reroute the network connection based on which links reach bandwidth capacity.
19. The system of claim 15 , wherein the changed total allowed bandwidth is also based on historical slice throughput.
20. The system of claim 15 , wherein the changed total allowed bandwidth is further based on a number of slices passing through the network connection.