Network quality of service management

• US 20050195741A1
• Filed: 04/02/2004
• Published: 09/08/2005
• Est. Priority Date: 03/03/2004
• Status: Active Grant

• Current Assignee: WSOU Investments, LLC (WSOU Holdings, LLC)
• Original Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Inventors: Wang, Yung-Terng, Doshi, Bharat T., Samadi, Behrokh
• Granted Patent: US 7,609,637 B2
• Time in Patent Office: Days
• US Class Current: 370/230
• CPC Class Codes: H04L 45/34 Source routing
• CPC Class Codes: H04L 45/50 using label swapping, e.g. multi-protocol label switch [MPLS]
• CPC Class Codes: H04L 47/78 Architectures of resource allocation
• CPC Class Codes: H04L 47/781 Centralised allocation of resources
• CPC Class Codes: H04L 47/782 Hierarchical allocation of resources, e.g. involving a hierarchy of local and centralised entities
• CPC Class Codes: H04L 47/783 Distributed allocation of resources, e.g. bandwidth brokers
• CPC Class Codes: H04L 47/805 QOS or priority aware
• CPC Class Codes: H04L 47/825 Involving tunnels, e.g. MPLS

Abstract

An architecture for quality-of-service (QoS) management creates a logical circuit-switched network within a packet network to support QoS-sensitive demands levied on the network. This QoS-managed network can serve to interwork, e.g., a PSTN with VoIP networks. The architecture can include a connection resource manager (CRM), which oversees bandwidth availability and demand admission/rejection on dynamically provisioned virtual trunk groups (VTGs) within the packet network, and a transport bandwidth controller (TBC). The VTGs serve to transport QoS-sensitive demands across the packet network. The TBC serves the CRM by providing an interface to routers and/or OAM systems of the packet network to size VTGs to meet QoS requirements. Media switches located at the packet network borders serve to mux/demux the demands into/from VTGs. CRMs and TBCs can be implemented as centralized, distributed, or hierarchical, and flat and aggregated variants of the architecture are supported. VTGs can be implemented using MPLS LSPs, VPNs, or source-based routing.

Claims

1. A packet network comprising:

a set of packet-switching nodes inconnected by communication links; and

at least one connection resource manager (CRM) adapted to overlay a logical circuit-switched network onto at least a subset of the set of packet-switching nodes and corresponding communication links, which logical circuit-switched network provides transport for quality-of-service (QoS)-sensitive demands within the packet network, wherein the logical circuit-switched network provides a guaranteed QoS to demands it carries.

2. The invention of claim 1, wherein the logical circuit-switched network comprises one or more virtual trunk groups (VTGs).

3. The invention of claim 2, wherein each VTG corresponds to a path within the packet network interconnecting two of the nodes.

4. The invention of claim 3, wherein each VTG has at least one of a guaranteed bandwidth, a guaranteed upper bound on jitter, a guaranteed probabilistic upper bound on jitter, a guaranteed upper bound on packet delay, a guaranteed probabilistic upper bound on packet delay, and a guaranteed upper bound on packet loss ratio.

5. The invention of claim 3, wherein each VTG has a guaranteed bandwidth, a guaranteed upper bound on jitter, a guaranteed upper bound on packet delay, and a guaranteed upper bound on packet loss.

6. The invention of claim 3, wherein each VTG has a guaranteed bandwidth, a guaranteed probabilistic upper bound on jitter, a guaranteed probabilistic upper bound on packet delay, and a guaranteed upper bound on packet loss ratio.

7. The invention of claim 3, wherein the network is an Internet protocol (IP) network and at least one of the VTGs is implemented using at least one of a virtual private network (VPN) technology and a source-based routing technology.

8. The invention of claim 3, wherein the network is a multiprotocol label switching (MPLS) enhanced IP network and each VTG is a label-switched path (LSP).

9. The invention of claim 2, wherein the CRM is adapted to maintain information about each VTG, wherein the information comprises available bandwidth in the VTG.

10. The invention of claim 9, wherein the CRM is adapted to:

receive a request for a new QoS-sensitive demand in the packet network, which request specifies a demand bandwidth for the new QoS-sensitive demand;

select at least one VTG to carry the new QoS-sensitive demand as a function of the available bandwidth for the selected VTG; and

reduce the available bandwidth for at least the selected VTG based on the demand bandwidth.

11. The invention of claim 10, wherein the selecting comprises comparing the demand bandwidth with the available bandwidth of at least two different VTGs.

12. The invention of claim 10, wherein a series of two or more concatenated VTGs are selected to carry the QoS-sensitive demand.

13. The invention of claim 10, wherein selecting the at least one VTG involves at least one of provisioning a new VTG in the packet network or provisioning more capacity on an existing VTG in the network.

14. The invention of claim 9, wherein the CRM is adapted to perform a comparison between the available bandwidth of at least one VTG with a lower bandwidth bound for the VTG and request additional capacity for the VTG as a function of at least the result of the comparison.

15. The invention of claim 9, wherein the CRM is adapted to perform a comparison between the available bandwidth of at least one VTG with an upper bandwidth bound for the VTG and request a release of capacity for the VTG as a function of at least the result of the comparison.

16. The invention of claim 9, wherein the CRM is adapted to adjust an upper bandwidth bound and a lower bandwidth bound for at least one VTG as a function of QoS requirements related to new demand blocking rates.

17. The invention of claim 2, further comprising a plurality of media switches (MSWs), each MSW operatively coupled between a plurality of input VTGs and a plurality of output VTGs and each MSW is adapted to multiplex demands from two or more of the plurality of input VTGs to one or more of the plurality of output VTGs.

18. The invention of claim 2, further comprising a plurality of media switches (MSWs), each MSW operatively coupled between a plurality of input VTGs and a plurality of output VTGs and each MSW is adapted to demultiplex demands from one or more of the plurality of input VTGs to two or more of the plurality of output VTGs.

19. The invention of claim 17, wherein at least one MSW is operatively coupled to at least a first and a second input VTG and at least one output VTG and the at least one MSW is adapted to multiplex demands from the first and second input VTGs into the output VTG, wherein:

the first input VTG couples a first media gateway (MGW) to the MSW;

the second input VTG couples a second MGW to the MSW; and

the output VTG couples the MSW to one other MSW in the packet network.

20. The invention of claim 19, wherein the at least one CRM comprises a first CRM and a second CRM, wherein:

the first CRM manages resources associated with at least the first input VTG; and

the second CRM manages resources associated with at least the output VTG.

23. The invention of claim 2, further comprising a plurality of media switches (MSWs), each MSW operatively coupled between a plurality of input VTGs and a plurality of output VTGs and adapted to switch at least one demand from at least one of the plurality of input VTGs to at least one of the plurality of output VTGs.

24. The invention of claim 1, wherein the CRM is adapted to communicate with the packet network to dynamically reprovision network resources in support of the QoS-sensitive demands.

25. The invention of claim 1, wherein the CRM is implemented using one or more computing elements wherein at least one of the computing elements is located remote from at least one other of the computing elements.

26. The invention of claim 1, wherein the packet network is at least one of a voice-over-Internet-Protocol (VoIP) network, an ATM network, and a frame relay network.

27. The invention of claim 1, further comprising a transport bandwidth controller (TBC) adapted to support communication between an application services network and the packet network.

28. The invention of claim 27, wherein the CRM processes application service requests from the application services network and passes the processed application service requests to the TBC.

29. The invention of claim 27, wherein the application services network is a voice services network.

30. The invention of claim 27, wherein the CRM is adapted to dynamically manage bandwidth on VTGs by communicating bandwidth allocation and/or deallocation requests to the packet network via the TBC.

31. The invention of claim 1, wherein the logical circuit-switched network is adapted to span at least two domains, wherein each domain is of a particular type and each domain type is at least one of the domain types of the group consisting of technology domains and administrative domains.

32. The invention of claim 1, wherein the logical circuit-switched network is adapted to span at least two domains and the at least one CRM comprises at least two border CRMs (B-CRMs), wherein the B-CRMs coordinate management of network resources between the domains.

21. The invention of 17, wherein two or more of the MSWs are core MSWs (C-MSW), wherein C-MSWs can include all of the functionality of MSWs and at least one of the C-MSWs is operatively coupled to at least a first and a second input VTG and at least one output VTG is adapted to multiplex demands from at least the first and second input VTGs into at least the at least one output VTG, wherein:

the first input VTG couples a first MSW to the C-MSW;

the second input VTG couples either a second MSW or a MGW to the C-MSW; and

the output VTG couples the C-MSW to at least one other C-MSW in the packet network.

22. The invention of claim 21, wherein the at least one CRM comprises a first CRM and a second CRM, wherein:

the first CRM manages resources associated with at least the first input VTG; and

the second CRM manages resources associated with at least the output VTG.

33. A method for providing quality-of-service (QoS) quarantees across a packet network, which packet network includes a set of packet-switching nodes interconnected by communication links, the method comprising overlaying a logical circuit-switched network onto at least a subset of the set of packet-switching nodes and the communications links, which logical circuit-switched network provides transport for QoS-sensitive demands within the packet network, wherein the logical circuit-switched network provides a guaranteed QoS to demands it carries.

34. A connection resource manager (CRM) adapted to:

receive a request for a new quality-of-service (QoS) sensitive demand in a packet network, which request specifies a demand bandwidth for the new QoS-sensitive demand;

select at least one virtual trunk group (VTG) to carry the new QoS-sensitive demand as a function of the available bandwidth for the at least one VTG, wherein each VTG corresponds to a path within the packet network interconnecting two or more packet-switching nodes; and

reduce the available bandwidth for the at least one VTG based on the demand bandwidth.

35. The invention of claim 34, wherein the CRM is adapted to communicate with the packet network to dynamically reprovision network resources in support of the QoS-sensitive demands.

36. The invention of claim 34, wherein the CRM is implemented using one or more computing elements wherein at least one of the computing elements is remotely located from at least one other of the computing elements.

37. The invention of claim 34, wherein the CRM is adapted to maintain information about each VTG, wherein the information comprises available bandwidth in the VTG.

38. A method for overlaying a virtual circuit-switched network onto a packet network, the method comprising:

receiving, at a packet-switching node of the packet network, a request to establish a virtual trunk group (VTG) through the packet-switching node, the request identifying at least a guaranteed bandwidth for the VTG, an input port associated with the VTG, and an output port associated with the VTG;

reserving, at the packet-switching node, the guaranteed bandwidth between the input and output ports associated with the VTG; and

receiving, at the packet-switching node, a request to change the guaranteed bandwidth for the VTG.

39. A packet-switching node in a packet network, the packet-switching node adapted to:

receive a request to establish a virtual trunk group (VTG) through the packet-switching node, the request identifying at least a guaranteed bandwidth for the VTG, an input port associated with the VTG, and an output port associated with the VTG;

reserve the guaranteed bandwidth between the input and output ports associated with the VTG; and

receive, at the packet-switching node, a request to change the guaranteed bandwidth for the VTG.

40. A method for overlaying a virtual circuit-switched network onto a packet network, the method comprising:

means for receiving, at a packet-switching node of the packet network, a request to establish a virtual trunk group (VTG) through the packet-switching node, the request identifying at least a guaranteed bandwidth for the VTG, an input port associated with the VTG, and an output port associated with the VTG;

means for reserving, at the packet-switching node, the guaranteed bandwidth between the input and output ports associated with the VTG; and

means for receiving, at the packet-switching node, a request to change the guaranteed bandwidth for the VTG.