Peer-influenced aggregate member selection
View Patent ↗Embodiments of the present invention provide the ability to allow a peer or remote device to influence aggregate member selection in an aggregate group of members for a data flow. In embodiments, local-side weights and remote-side weights may be assigned to each member in an aggregate group of members. Using these weights, a global value for each member in the aggregate group may be calculated. Given these global values, a member from the aggregate group of members may be selected for egress of the new flow based upon the global values.
1. An information handling system (IHS) comprising:
a plurality of ports, at least one of the plurality of ports being configured to receive a new flow and at least two or more of the ports forming an aggregate group of members, each member being potentially used as an egress for the new flow from the IHS to one or more remote devices;
one or more processors that are communicatively coupled to the plurality of I/O ports; and
a memory that is communicatively coupled to the one or more processors and stores one or more sequences of instructions, which when executed by one or more processors causes steps to be performed comprising:
assigning an IHS-side weight to each member in the aggregate group of members;
assigning a remote-side weight to each member in the aggregate group of members, a remote-side weight being obtained from a remote device corresponding with the member;
calculating a global value for each member in the aggregate group of members using the IHS-side weights and the remote-side weights; and
selecting for egress of the new flow a member from the aggregate group of members based upon the global values.
2. The information handling system of claim 1 further comprising:
receiving IHS-remote preference weights that indicate a weighting preference for the an IHS-side weights and the remote-side weights; and
wherein the step of calculating a global value for each member in the aggregate group of members using the IHS-side weights and the remote-side weights also comprising using the IHS-remote preference weights.
3. The information handling system of claim 2 wherein the IHS-remote preference weights are user selected.
4. The information handling system of claim 2 wherein the step of assigning an IHS-side weight to each member in the aggregate group of members comprises basing the IHS-side weight for a member upon one or more metrics related to utilization of the member at the IHS.
5. The information handling system of claim 2 wherein the step of assigning a remote-side weight to each member in the aggregate group of members comprises basing the remote-side weight for the member upon one or more metrics related to utilization of the member at the remote device.
6. The information handling system of claim 5 wherein the remote-side weights are received from the one or more remote devices.
7. The information handling system of claim 1 further comprising:
re-calculating the global values at different instances of time.
8. The information handling system of claim 7 further comprises, prior to the step of re-calculating the global values:
updating at least some of the IHS-side weights and the remote-side weights.
9. A processor-implemented method assigning a member from an aggregate group of members as an egress for a new flow received at a local device to a remote device, the method comprising:
assigning a local-side weight to each member in the aggregate group of members;
assigning a remote-side weight to each member in the aggregate group of members, a remote-side weight being obtained from a remote device corresponding with the member;
calculating a global value for each member in the aggregate group of members using the local-side weights and the remote-side weights; and
selecting for egress of the new flow a member from the aggregate group of members based upon the global values.
10. The processor-implemented method of claim 9 further comprising:
receiving local-remote preference weights that indicate a weighting preference for the an local-side weights and the remote-side weights; and
wherein the step of calculating a global value for each member in the aggregate group of members using the local-side weights and the remote-side weights also comprising using the local-remote preference weights.
11. The processor-implemented method of claim 10 wherein the local-remote preference weights are user selected.
12. The processor-implemented method of claim 10 wherein the step of assigning an local-side weight to each member in the aggregate group of members comprises basing the local-side weight for a member upon one or more metrics related to utilization of the member at the local device.
13. The processor-implemented method of claim 10 wherein the step of assigning a remote-side weight to each member in the aggregate group of members comprises basing the remote-side weight for the member upon one or more metrics related to utilization of the member at the remote device.
14. The processor-implemented method of claim 13 wherein the remote-side weights are received from the one or more remote devices.
15. The processor-implemented method of claim 9 further comprising:
re-calculating the global values at different instances of time.
16. The processor-implemented method of claim 15 further comprises, prior to the step of re-calculating the global values:
updating at least some of the local-side weights and the remote-side weights.
17. An information handling system (IHS) comprising:
a plurality of ports, at least one of the plurality of ports being configured to receive a new flow and at least two or more of the ports forming an aggregate group of members, each member being potentially used as an egress for the new flow from the IHS to one or more remote devices;
one or more processors that are communicatively coupled to the plurality of I/O ports; and
a memory that is communicatively coupled to the one or more processors and stores one or more sequences of instructions, which when executed by one or more processors causes steps to be performed comprising:
assigning an IHS-side weight to each member in the aggregate group of members using one or more IHS-side egress accounting metrics;
assigning a remote-side weight to each member in the aggregate group of members using one or more ingress accounting metrics, a remote-side weight being obtained from a remote device corresponding with the member;
calculating a global value for each member in the aggregate group of members using the IHS-side weights and the remote-side weights; and
selecting for egress of the new flow a member from the aggregate group of members based upon the global values.
18. The information handling system of claim 17 further comprising:
receiving IHS-remote preference weights that indicate a weighting preference for the an IHS-side weights and the remote-side weights; and
wherein the step of calculating a global value for each member in the aggregate group of members using the IHS-side weights and the remote-side weights also comprising using the IHS-remote preference weights.
19. The information handling system of claim 18 wherein the IHS-remote preference weights are user selected.
20. The information handling system of claim 17 further comprising:
updating at least some of the IHS-side weights and the remote-side weights; and
re-calculating the global values.