摘要:
The subject matter described herein includes methods, systems, and computer program products for load balanced and symmetric SPF path computations for VoIP traffic engineering. One method includes advertising an output interface cost in a first direction over a first network segment between a source IP device and a destination IP device that is different from an output interface cost in advertised in a second direction over the first network segment. A plurality of available network paths between the source IP device and the destination IP device is identified. A path cost is calculated for each of the available network paths from the source IP device to the destination IP device. Calculating a path cost includes, for paths that include the first network segment, substituting the output interface cost advertised in the first direction with the output interface cost advertised in the second direction. The path costs are ranked based on the calculated costs. A lowest cost path is selected from the ranked paths. Both directions of a media session are assigned to the lowest cost path.
摘要:
Methods, systems, and computer program products for multipath Shortest-Path-First (SPF) computations and distance-based interface selection for VoIP traffic are disclosed. According to one method, a multi-path router instance associated with a plurality of network interfaces in a source IP device is provided. A cost is assigned to each of a plurality of internal segments between the multi-path router instance and the network interfaces associated with the multi-path router instance. An aggregate cost is calculated for each of a plurality of traffic paths originating at the multi-path router instance in the source IP device and extending through each of the network interfaces associated with the multi-path router instance to a destination IP device in the network. A list of IP paths is generated, and the paths in the list are ranked based on the calculated cost. Traffic is sent from the source IP device over at least one of the paths in the list. The path costs and rankings are updated in response to segment cost changes.
摘要:
Methods and systems for per-session dynamic management of media gateway resources are disclosed. According to one method, the logical and physical resources in a media gateway are divided and dynamically managed at the Transport Layer (i.e. OSI Layer 4), which results in finer granularity than managing such resources statically at the Data Link Layer (i.e. OSI Layer 2) or Network Layer (i.e. OSI Layer 3). Voice-processing resources provided by voice server cards may be pooled into a common pool available to all external networks. For each new call/session, the dynamic resource manager of the media gateway dynamically allocates a voice chip from the pooled voice processing resources, and assigns a logical resource identifier (e.g. a local IP and local UDP pair) to the session. When a network interface card receives incoming voice packets, it checks the destination IP and UDP and optionally the source IP and UDP to find out, and forward voice packets to, the voice chip assigned to the session.
摘要:
Methods and systems for providing voice over IP traffic engineering and path resilience using a network-aware media gateway are provided. In a media gateway, voice over IP hosts are assigned a first set of IP addresses. Network interfaces in the media gateway are assigned a second set of IP addresses that differ from the first set of IP addresses. Per-interface routers advertise reachability information from at least one of the voice over IP hosts via multiple interfaces and participate in network routing protocols to generate per interface routing tables. Voice over IP path lists may be generated based on the per interface routing tables. Internal costs may be assigned to the associations between the voice over IP hosts and the interfaces based on traffic engineering criteria.
摘要:
Method, systems, and computer program products for voice over IP (VoIP) traffic engineering and path resilience using media gateway and associated next-hop routers are provided. In one example, a media gateway includes multiple media processing hosts and a multiple network interfaces separate from the media processing hosts. At least one of the media processing hosts is reachable via at least two of the network interfaces. Next-hop routers associated with the network interfaces advertise reachability information regarding the media processing hosts to routers in the network. The next-hop routers also participate in network routing protocols to form multiple paths between the media gateway and a remote media gateway. Costs may be assigned to associations between the media processing hosts and the next-hop routers for traffic engineering purposes.
摘要:
Methods, systems, and computer program products for throttling network address translation (NAT) learning traffic in a voice over IP device are disclosed. According to one method, a plurality of media packets associated with a media session are received at a voice over IP device. A NAT learning throttling filter is applied to select the subset of the packets to be used for NAT learning and thereby limit the number of received media packets to be used for NAT learning. NAT learning is performed for the session using the packets selected by the NAT learning throttling filter.
摘要:
Methods, systems, and computer program products for implementing link redundancy in a media gateway are provided according to one method, a media gateway link protection group is provisioned to associate with a common VLAN a common virtual local area network (VLAN) primary and secondary links associated with at least one line card in a media gateway. The primary and secondary links are connected using a cross connection between ports associated with the at least one line card. At run time, traffic is automatically bridged between the primary and secondary links using the common VLAN and the cross connection.
摘要:
An apparatus that includes a packet data ingress, a packet data egress, and a packet data switching matrix configured to switch packet data from the packet data ingress to the packet data egress. The apparatus may further include a packet data jitter buffer and a packet data splitter, wherein the packet data splitter interposes the packet data ingress and the packet data switching matrix and is configured to multicast the packet data to the packet data switching matrix and the jitter buffer.
摘要:
An apparatus including a plurality of packet processors each included in one of a plurality of voice-over-internet-protocol (VoIP) network interfaces. Each of the plurality of packet processors is configured to cache a latest version of realtime transport control protocol (RTCP) report data by discarding an older version of the RTCP report data. The RTCP report data includes at least one of RTCP sender report data and RTCP receiver report data. The apparatus also includes a packet data switching matrix configured to switch packet data between ones of the plurality of VoIP network interfaces. A central processor of the apparatus is configured to generate a final session detail record upon the termination of a VoIP-session by selecting RTCP session-concluding report data from a plurality of RTCP final report data each cached by a corresponding one of the plurality of packet processors.
摘要:
An apparatus that includes a packet data ingress, a packet data egress, and a packet data switching matrix configured to switch packet data from the packet data ingress to the packet data egress. The apparatus may further include a packet data jitter buffer and a packet data splitter, wherein the packet data splitter interposes the packet data ingress and the packet data switching matrix and is configured to multicast the packet data to the packet data switching matrix and the jitter buffer.