Method and system for bandwidth allocation tracking in a packet data network
View Patent ↗In a metropolitan area network, a method and system for maintaining an accurate total of the amount of allocated bandwidth on the network. A plurality of incoming packets are assigned to a respective plurality of queues of a metropolitan area network switch. Using a fair arbitration scheme, the respective queues are configured to empty at a specified output rate. A finish time for each respective queue is computed, the finish time describing a time at which the respective queue will be emptied using the output rate. The plurality of queues are grouped into multiple groups in accordance with their respective finish times. The earliest group includes the reserved rates of those queues having a finish time indicating an empty condition at a first time increment. The second earliest group includes the reserved rates of those queues having a finish time indicating an empty condition at a second time increment later than the first time increment, and so on. The amount of allocated bandwidth on the network is determined by tracking the sum of the reserved rates of all the multiple groups. The first time increment, second time increment, and the like are indexed with respect to a schedule clock. The earliest group thus indicates those queues that will have an empty condition at a next time increment of the schedule clock. The determination of the amount of allocated bandwidth can be accomplished in real time, thereby allowing the efficient allocation of unallocated bandwidth in real time.
1. In a packet data network, a method for maintaining an accurate total of the amount of allocated bandwidth on the network, the method comprising the steps of:
a) assigning a plurality of incoming packets to a respective plurality of queues, the respective queues configured to empty at an output rate
b) computing a finish time for each respective queue, the finish time describing a time at which the respective queue will be emptied using the output rate
c) grouping the plurality of queues into at least a first group and a second group, wherein the first group includes those queues having a finish time indicating an empty condition at a first time increment and the second group includes those queues having a finish time indicating an empty condition at a second time increment later than the first time increment; and
d) determining the amount of allocated bandwidth by totaling number of queues in the first group and the second group.
2. The method of claim 1 wherein a respective weight is associated with each of the plurality of queues and an allocated weight for the amount of allocated bandwidth is determined by totaling the respective weights of the queues in the first group and the second group.
3. The method of claim 1 wherein the first time increment and the second time increment are indexed with respect to a schedule clock and the first group indicates those queues that will have an empty condition at a next increment of the schedule clock.
4. The method of claim 3 wherein the schedule clock is incremented in accordance with a relative amount of congestion measured at a node.
5. The method of claim 1 wherein the totaling from step d) is maintained in a database with corresponding weights for the number of queues.
6. The method of claim 1 wherein a new finish time is computed for each respective queue when a new packet is received by the respective queue.
7. The method of claim 1 wherein the amount of allocated bandwidth is determined in real time.
8. The method of claim 1 wherein the amount of allocated bandwidth is used to determine an amount of unallocated bandwidth, thereby allowing the allocation of the unallocated bandwidth while maintaining a quality of service of the allocated bandwidth.
9. A system for distributing bandwidth in a packet data network while maintaining an accurate total of the amount of allocated bandwidth on the network, comprising:
a respective plurality of queues configured to receive a respective plurality of incoming packets of data flows, the queues included in an MPS (metro packet switch) coupled to the network;
a schedule clock configured to increment in relation to a measured congestion at a link; and
a database for grouping the plurality of queues into at least a first group and a second group, wherein the first group includes those queues having a finish time indicating an empty condition at a first time increment of the schedule clock and the second group includes those queues having a finish time indicating an empty condition at a second time increment of the schedule clock later than the first time increment, the MPS configured to determine the amount of allocated bandwidth by totaling the number of queues in the first group and the second group.
10. The system of claim 9 wherein a respective weight is associated with each of the plurality of queues and an allocated weight for the amount of allocated bandwidth is determined by totaling the respective allocated weights of the queues in the first group and the second group.
11. The system of claim 9 wherein a result the totaling of the queues is maintained in the database with corresponding weights for the queues.
12. The system of claim 9 wherein a new finish time is computed for each respective queue when a new packet is received by the respective queue.
13. The system of claim 9 wherein the amount of allocated bandwidth is determined in real time.
14. The system of claim 9 wherein the amount of allocated bandwidth is used to determine an amount of unallocated bandwidth, thereby allowing the allocation of the unallocated bandwidth while maintaining a quality of service of the allocated bandwidth.
15. The system of claim 9 wherein the amount of unallocated bandwidth is described as a single term, and the term is broadcast to a plurality of other nodes so that the other nodes can shape their input traffic.
16. The system of claim 9 wherein the database is implemented using pairs of counters for each group, each pair having a first counter for the number of queues within the group and a second counter for the respective weights of the queues with the group.
17. The system of claim 9 wherein the queues are virtual queues, wherein each of the virtual queues keeps track of a backlog of a corresponding data flow without physically buffering the data flow.
18. The system of claim 9 wherein the network is a ring topology metropolitan area network.
19. The system of claim 18 wherein the ring topology metropolitan area network comprises an Ethernet communications channel.