Abstract: A multiprocessor computer system comprises a sending processor node and a receiving processor node. The sending processor node is operable to send packets comprising part of a message to a receiver, and to send a message complete packet after all packets in the message are sent. The message complete packet includes an indicator of the number of packets in the message, and the message is recognized as complete in the receiver once the number of packets indicated in the message complete packet have been received for the message. The sender tracks acknowledgment from the receiver of receipt of the sent packets; and notifies the receiver when it has received all packets comprising a part of the message.

Abstract: A switching network includes an upper tier having a master switch and a lower tier including a plurality of lower tier entities. The master switch, which has a plurality of ports each coupled to a respective lower tier entity, implements on each of the ports a plurality of virtual ports each corresponding to a respective one of a plurality of remote physical interfaces (RPIs) at the lower tier entity coupled to that port. Data traffic communicated between the master switch and RPIs is queued within virtual ports that correspond to the RPIs with which the data traffic is communicated. The master switch applies data handling to the data traffic in accordance with a control policy based at least upon the virtual port in which the data traffic is queued, such that the master switch applies different policies to data traffic queued to two virtual ports on the same port of the master switch.

Abstract: A method and structure for switching data is provided. An output port scheduler obtains state information of VOQs and available state information of input port data channels and output port buffers. The output port scheduler sends scheduling request information to a FIC of an input port whose input port data channel is ready in input ports corresponding to non-empty VOQs pointing to an output port. After receiving the scheduling request information sent by the output port schedulers, the FIC of the selected input port selects to respond to a scheduling request of one output port scheduler, and sends the VOQ pointing to the output port in the selected input port to the output port buffer. The output port scheduler schedules the VOQ received by the output port buffer out of a switch chip. Buffer resources are saved and the switching performance is improved.

Abstract: A relay device relays between two TCP communication items of a LAN side and a WAN side. When a line bandwidth of the WAN side is smaller than a line bandwidth of the LAN side, buffer overflow in a LAN side reception buffer and a WAN side transmission buffer of the relay device is prevented, and a connection is prevented from being forced to be canceled. A value of a reception window size (rwnd) described in an ACK packet returned to a transmission terminal of the LAN side is controlled based on a transmission throughput, a discarding rate, and an RTT measured in TCP communication of the WAN side, and a total size of unarranged data and a size of arranged data in a reception buffer of the LAN side and a size of untransmitted data and ACK awaiting data in a transmission buffer of the WAN side.

Abstract: A network device for processing data includes at least one ingress module for performing switching functions on incoming data, a memory management unit for storing the incoming data and at least one egress module for transmitting the incoming data to at least one egress port. The at least one egress module includes an egress scheduling module and multiple queues per each of the at least one egress port. Each of the multiple queues serve data attributable to a class of service, and the egress scheduling module is configured to service a minimum bandwidth requirement for each of the multiple queues and then to service the multiple queues to allow for transmission of a maximum allowable bandwidth through a weighting of each of the multiple queues.

Abstract: A bridge routing module can be incorporated into a closed network fabric, such as a vehicular network. The bridge routing module includes an interface circuit to be coupled to other elements of the closed network fabric, for example other bridge routing modules or switch modules. The bridge routing module includes memory to store information associating packet content types with packet routing parameters, among other things. A processing module included in the bridge routing module analyzes packets to identify the type of content carried by the packets, and determines packet routing parameters based on the packet's content type. Ingress and egress of the packet are controlled in accordance with the packet routing parameters determined by the processing module.

Abstract: A network switch including a plurality of ports; a memory, and a queue controller. The queue controller is configured to: maintain a list of pointers to a first plurality of buffers in the memory; of the first plurality of buffers, selectively allocate a first buffer to a first port of the plurality of ports; in response to i) the first port receiving a first frame of data, ii) the first buffer being allocated to the first port, and iii) the first frame being stored in the memory, remove the pointer to the first buffer from the list of pointers; transfer, to an output queue associated with a second port of the plurality of ports, the pointer to the first buffer; and in response to the first frame of data being sent from the second port, add the pointer to the first buffer back to the list of pointers.

Abstract: A method includes receiving flits forwarded from an upstream router into a first input virtual channel (VC) associated with an input port. The flits are associated with packets originated from a first Intellectual Property (IP) core and forwarded to a second IP core. The flits are stored in a VC storage associated with the first input VC. The method further includes performing link width conversion based on a width of the flits being different from a width of an output port. Link width conversion includes accumulation of the flits when the width of the output port is wider and unpacking of the flits when the width of the output port is narrower. Credits are generated based on the flits being forwarded from the first input VC to the output port. The credits are sent to the upstream router to enable receiving more flits from the upstream router.

Abstract: An audio-on-demand communication system provides real-time playback of audio data transferred via telephone lines or other communication links. One or more audio servers include memory banks which store compressed audio data. At the request of a user at a subscriber PC, an audio server transmits the compressed audio data over the communication link to the subscriber PC. The subscriber PC receives and decompresses the transmitted audio data in less than real-time using only the processing power of the CPU within the subscriber PC. According to one aspect of the present invention, high quality audio data compressed according to lossless compression techniques is transmitted together with normal quality audio data. According to another aspect of the present invention, metadata, or extra data, such as text, captions, still images, etc., is transmitted with audio data and is simultaneously displayed with corresponding audio data.

Abstract: A method and apparatus for processing message is described. In one embodiment, an application programming interface is configured for receiving and sending messages. A multiplexer receives messages from different servers. A service name is coupled to each message with the corresponding destination service. A single shared channel is formed. The messages are processed over the single shared channel.

Abstract: Disclosed are methods, systems, paradigms and structures for processing data packets in a communication network by a multi-core network processor. The network processor includes a plurality of multi-threaded core processors and special purpose processors for processing the data packets atomically, and in parallel. An ingress module of the network processor stores the incoming data packets in the memory and adds them to an input queue. The network processor processes a data packet by performing a set of network operations on the data packet in a single thread of a core processor. The special purpose processors perform a subset of the set of network operations on the data packet atomically. An egress module retrieves the processed data packets from a plurality of output queues based on a quality of service (QoS) associated with the output queues, and forwards the data packets towards their destination addresses.

Abstract: A method and apparatus for switching a data packet between a source and destination in a network. The data packet includes a header portion and a data portion. The header portion includes routing information for the data packet. The method includes defining a data path in the router comprising a path through the router along which the data portion of the data packet travels and defining a control path comprising a path through the router along which routing information from the header portion travels. The method includes separating the data path and control path in the router such that the routing information can be separated from the data portion allowing for the separate processing of each in the router. The data portion can be stored in a global memory while routing decisions are made on the routing information in the control path.

Abstract: An approach for providing flow control in a radio communication system is disclosed. A request from a non-satellite system specific side of a transport interface is made to a system specific side of the transport interface for a flow control allocation that specifies an amount of data to be stored in a queue of the system specific side of the transport interface. The system specific side supports a signaling function that is based on a transmission characteristic of the radio communication system. The flow control allocation is generated based upon availability of the queue, wherein the destination address is a link layer address of the satellite communication system. This arrangement has particular applicability to a satellite network (e.g., Very Small Aperture Terminal (VSAT) network) that provides data communication services.

Abstract: A network interface adapter includes a network interface and a client interface, for coupling to a client device so as to receive from the client device work requests to send messages over the network using a plurality of transport service instances. Message processing circuitry, coupled between the network interface and the client interface, includes an execution unit, which generates the messages in response to the work requests and passes the messages to the network interface to be sent over the network. A memory stores records of the messages that have been generated by the execution unit in respective lists according to the transport service instances with which the messages are associated. A completion unit receives the records from the memory and, responsive thereto, reports to the client device upon completion of the messages.

Abstract: Embodiments of a system that includes a switch and a buffer-management technique for storing signals in the system are described. In this system, data cells are dynamically assigned from a host buffer to at least a subset of switch-ingress buffers in the switch based at least in part on the occupancy of the switch-ingress buffers. This buffer-management technique may reduce the number of switch-ingress buffers relative to the number of input and output ports to the switch, which in turn may overcome the limitations posed by the amount of memory available on chips, thereby facilitating large switches.

Abstract: In general, techniques are described for measuring packet data unit (PDU) loss in a L2 virtual private network (L2VPN) service, such as a VPLS instance. In one example of the techniques, provider edge (PE) routers that participate in the L2VPN measure known unicast and multicast PDU traffic at the service endpoints for the instance to determine unicast PDU loss within the service provider network. As the routers learn the outbound service (i.e., core-facing) interfaces and outbound local (i.e., customer-facing) interfaces for L2 addresses of customer devices that issue packets to the VPLS instance, the routers establish respective unicast transmit and receipt counters for the service endpoints that serve the customer devices. In another example, PE routers that participate in the L2VPN measure multicast PDU traffic at the service endpoints for the instance and account for internal replication by intermediate service nodes to determine multicast PDU loss within the service.

Abstract: A mechanism for combining plurality of point-to-point data channels to provide a high-bandwidth data channel having an aggregated bandwidth equivalent to the sum of the bandwidths of the data channels used is provided. A mechanism for scattering segments of incoming data packets, called data chunks, among available point-to-point data channel interfaces is further provided. A decision as to the data channel interface over which to send a data chunk to can be made by examining a fullness status of a FIFO coupled to each interface. An identifier of a data channel on which to expect a subsequent data chunk can be provided in a control word associated with a present chunk of data. Using such information in control words, a receive-end interface can reassemble packets by looking to the control word in a currently processing data chunk to find a subsequent data chunk.

Abstract: The present invention provides switches and routers, preferably with fully-connected mesh fabrics, that transmit data through the switch fabric in variable-size data units. Variable-size data units allow switches and routers to provide throughputs close to hardware capabilities, eliminating the need for over-capacity hardware in the switch fabric and other components. Along with variably-size data units, preferred embodiments of this invention include scheduling methods that provide fair allocation of pre-determined bandwidths to different protocols, to different classes of service within protocols, and to different resources within the switch by use of certain weighted, fair scheduling methods. The switches and routers of this invention are particularly directed to multi-protocol, high-throughput communication applications, but may have wide applicability in systems generally where data packets are switched or routed.

Abstract: In one embodiment, the present invention includes a method for determining whether a packet received in an input/output (I/O) circuit of a node is destined for the node and if so, providing the packet to an egress queue of the I/O circuit and determining whether one or more packets are present in an ingress queue of the I/O circuit and if so, providing a selected packet to a first or second output register according to a global schedule that is independent of traffic flow. Other embodiments are described and claimed.

Abstract: Transmitting from a mobile terminal to a telecommunication network data stored in a plurality of queues, each queue having a respective transmission priority, includes setting the data in each of the queues to be either primary data or secondary data, or a combination of primary data and secondary data. The data may be transmitted from the queues in an order in dependence upon the priority of the queue and whether the data in that queue are primary data or secondary data. Resources for data transmission may be allocated such that the primary data of each of the queues are transmitted at a minimum predetermined rate and such that the secondary data of each of the queues are transmitted at a maximum predetermined rate, greater than the minimum predetermined rate.

Abstract: A system and method for implementing a VM to identify a data packet for transmission, the data packet including a QoS the data packet is to receive as compared to another QoS that another data packet is to receive. The system and method further includes a SNIC to pull the data packet from the VM based upon the QoS the data packet is to receive. The system and method may also include a link scheduler module to transmit the data packet based upon the QoS the data packet is to receive. The system and method may also include a receiver to receive a management instruction from a network management device, the management instruction to dictate the QoS the data packet is to receive based upon a SLA.

Abstract: A switch includes a reserved pool of buffers in a shared memory. The reserved pool of buffers is reserved for exclusive use by an egress port. The switch includes pool select logic which selects a free buffer from the reserved pool for storing data received from an ingress port to be forwarded to the egress port. The shared memory also includes a shared pool of buffers. The shared pool of buffers is shared by a plurality of egress ports. The pool select logic selects a free buffer in the shared pool upon detecting no free buffer in the reserved pool. The shared memory may also include a multicast pool of buffers. The multicast pool of buffers is shared by a plurality of egress ports. The pool select logic selects a free buffer in the multicast pool upon detecting an IP Multicast data packet received from an ingress port.

Abstract: A method and system for allocating exchange identifications (IDs) in a fiber channel switch for fiber channel aggregation. The method included determining a number (m) of N_ports present in a back end of the switch, and distributing available exchange IDs across the number (m) of present N_ports. Each exchange ID includes (j) bits and (n) bits are used to identify each of the present backend ports, where m?2n.

Abstract: One embodiment of the present invention provides a system that facilitates flow control of multi-path-switched data frames. During operation the system transmits from an ingress edge device data frames destined to an egress edge device across different switched paths based on queue status of a core switching device and queue status of the egress edge device. The egress edge device is separate from the core switching device.

Abstract: Disclosed are methods, systems, paradigms and structures for processing data packets in a communication network by a multi-core network processor. The network processor includes a plurality of multi-threaded core processors and special purpose processors for processing the data packets atomically, and in parallel. An ingress module of the network processor stores the incoming data packets in the memory and adds them to an input queue. The network processor processes a data packet by performing a set of network operations on the data packet in a single thread of a core processor. The special purpose processors perform a subset of the set of network operations on the data packet atomically. An egress module retrieves the processed data packets from a plurality of output queues based on a quality of service (QoS) associated with the output queues, and forwards the data packets towards their destination addresses.

Abstract: A switching network includes an upper tier and a lower tier including a plurality of lower tier entities. A master switch in the upper tier, which has a plurality of ports each coupled to a respective lower tier entity, implements on each of the ports a plurality of virtual ports each corresponding to a respective one of a plurality of remote physical interfaces (RPIs) at the lower tier entity coupled to that port. Data traffic communicated between the master switch and RPIs is queued within virtual ports that correspond to the RPIs on lower tier entities with which the data traffic is communicated. The master switch enforces priority-based flow control (PFC) on data traffic of a given virtual port by transmitting, to a lower tier entity on which a corresponding RPI resides, a PFC data frame specifying priorities for at least two different classes of data traffic communicated by the particular RPI.

Abstract: In one embodiment, the present invention includes a method for determining whether a packet received in an input/output (I/O) circuit of a node is destined for the node and if so, providing the packet to an egress queue of the I/O circuit and determining whether one or more packets are present in an ingress queue of the I/O circuit and if so, providing a selected packet to a first or second output register according to a global schedule that is independent of traffic flow. Other embodiments are described and claimed.

Abstract: A queue scheduling method and apparatus is disclosed in the embodiments of the present invention, the method comprises: one or more queues are indexed by using a first circulation link list; one or more queues are accessed respectively by using the front pointer of the first circulation link list, and the value acquired from subtracting a value of a unit to be scheduled at the head of the queue from a weight middle value of each queue is treated as the residual weight middle value of the queue; when the weight middle value of one queue in the first circulation link list is less than the unit to be scheduled at the head of the queue, the queue is deleted from the first circulation link list and the weight middle value is updated with the sum of a set weight value and the residual weight middle value of the queue; the queue deleted from the first circulation link list is linked with a second circulation link list.

Abstract: An Ethernet switch has at least one ingress/egress port which is operable in two modes, in a first mode as a GE port and in a second mode as a plurality of FE ports. The port has 8 MAC interfaces each of which is capable of receiving/transmitting FE packets, and at least one of the MAC interfaces can be configured to receive/transmit GE packets. Thus, the port has two modes of operation. The port further includes receive and transmit modules which receive GE and FE packets from, and transmit GE and FE packets to, the interfaces.

Abstract: A switching network includes an upper tier and a lower tier including a plurality of lower tier entities. A master switch in the upper tier, which has a plurality of ports each coupled to a respective lower tier entity, implements on each of the ports a plurality of virtual ports each corresponding to a respective one of a plurality of remote physical interfaces (RPIs) at the lower tier entity coupled to that port. Data traffic communicated between the master switch and RPIs is queued within virtual ports that correspond to the RPIs on lower tier entities with which the data traffic is communicated. The master switch enforces priority-based flow control (PFC) on data traffic of a given virtual port by transmitting, to a lower tier entity on which a corresponding RPI resides, a PFC data frame specifying priorities for at least two different classes of data traffic communicated by the particular RPI.

Abstract: A system determines a scheduling value based on a current length of a downstream queue in a network device. The system sends the scheduling value from the downstream queue to an upstream queue and schedules dequeuing of one or more data units, destined for the downstream queue, from the upstream queue based on the scheduling value.

Abstract: Described embodiments provide for arbitrating between nodes of scheduling hierarchy of a network processor. A traffic manager generates a tree scheduling hierarchy having a root scheduler and N scheduler levels. The network processor generates tasks corresponding to received packets. The traffic manager queues the received task in an associated queue of the scheduling hierarchy. The root scheduler performs smooth deficit weighted round robin (SDWRR) arbitration between each child node of the root scheduler. The SDWRR arbitration includes checking one or more status indicators of each child node of the given scheduler and selecting, based on the status indicators, a first active child node of the scheduler and updating the one or more status indicators corresponding to the selected child node. Thus, a task is scheduled for transmission by the traffic manager every cycle of the network processor.

Abstract: A device may include a first line card and a second line card. The first line card may include a memory including queues. In addition, the first line card may include a processor. The processor may identify, among the queues, a queue whose size is to be modified, change the size of the identified queue, receive a packet, insert a header cell associated with the packet in the identified queue, identify a second line card from which the packet is to be sent to another device in a network, remove the header cell from the identified queue, and forward the header cell to the second line card. The second line card may receive the header cell from the first line card, and send the packet to the other device in the network.

Abstract: A method of preparing data streams to facilitate seamless switching between such streams by a switching device to produce an output data stream without any switching artifacts. Bi-directional switching between any plurality of data streams is supported. The data streams are divided into segments, wherein the segments include synchronized starting points and end points. The data rate is increased before an end point of a segment, to create switch gaps between the segments. Increasing the data rate can include increasing a bandwidth of the plurality of data streams, for example by multiplexing, or compressing the data. The present invention can be used, for example, with MPEG or AC-3 encoded audio and MPEG encoded video segments that are multiplexed into MPEG-2 transport streams. Also included are specific methods for preparing MPEG video streams and multiplexing MPEG video with MPEG or AC-3 audio streams to allow a receiver to create seamless transitions between individually encoded segments.

Abstract: A method and apparatus for queue-ordering commands in multi-engines, multi-queues and/or multi-flows environment is provided. Commands from single/multiple queues and multi-flows are processed by multi-engines with different processing time and/or out of order, which breaks sequential order of commands from same input queue and commands are distributed across multiple engines' output buffer after processing. Processed commands are stored in dedicated command output buffer associated with each engine temporarily. The processed commands are re-ordered while writing out. Also commands can be scheduled to idle engines to achieve maximum throughput, thus utilizing the engines in an optimal manner.

Abstract: The invention provides a method for adding specific hardware on both receive and transmit sides that will hide to the software most of the effort related to buffer and pointers management. At initialization, a set of pointers and buffers is provided by software, in quantity large enough to support expected traffic. A Send Queue Replenisher (SQR) and Receive Queue Replenisher (RQR) hide RQ and SQ management to software. RQR and SQR fully monitor pointers queues and perform recirculation of pointers from transmit side to receive side.

Abstract: Embodiments of the invention are directed to multicasting packets in a system such as a data packet switch or router having a distributed architecture. A first device such as a line card receiving a packet that requires multicasting forwards the packet to a fabric switch where the packet is replicated to obtain one respective packet for each line card of the system. Each line card receives its respective packet from the fabric switch and further duplicates the packet to obtain a duplicate packet for each egress endpoint of a service associated with the packet that is eligible to receive such a duplicate packet. Replication and duplication of packets requiring multicasting performed in this manner efficiently uses bandwidth of the fabric switch and links connecting it to the line cards.

Abstract: A packet is classified into a class. A priority value is assigned to the packet wherein packets in a flow are assigned priorities according to some probability distribution within some band. A determination is made, at a network device for a highest latency class, whether a sum of queued packet sizes of previously received packets having an equal or smaller latency class than the packet and larger or equal priority than the packet is larger than a threshold value. When the sum is larger, the packet is dropped, otherwise a determination is made whether a latency class of the packet is less than the latency class of the network device. When the latency class is not less, the packet is stored in a queue for the latency class. When the latency class is less, then the process is repeated until the packet is dropped or stored in a queue.

Abstract: A mobile communications device with a wireless module and a controller module for performing a buffer status reporting procedure is provided. The wireless module performs wireless transmissions and receptions to and from a cellular station of a service network. The controller module receives machine type communication (MTC) data, determines whether to trigger a buffer status report (BSR) according to a comparison result of the data size of the MTC data and a threshold value when the MTC data arrives at an empty transmission buffer, and if so, transmits the BSR to the cellular station via the wireless module.

Abstract: Forwarding a flow in a network includes receiving the flow at a switch, determining an optimized priority queue level of the flow at the switch, and forwarding the flow via the switch using an optimized priority queue level of the flow at the switch. The flow passes through a plurality of switches, including the switch, in the network, and the optimized priority queue level of the flow at the switch is different from a priority queue level of the flow at a second switch of the plurality of switches. The second switch routes the flow at the second switch using the different priority queue level for the flow.

Abstract: An apparatus comprising a plurality of physical ingress ports configured to receive data, each data having a data type; a plurality of physical egress ports configured to transmit data; a memory configured to buffer data that has been received; a plurality of virtual routing devices, wherein each of the virtual routing devices is associated with a particular data type and each of the virtual routing devices is configured to: virtually buffer data associated with the respective data type, and regulate the quality of service provided to the respective data type; and a data manager configured to manage the receipt and transmission of data.

Abstract: A network device for processing data includes at least one ingress module for performing switching functions on incoming data, a memory management unit for storing the incoming data in a memory and at least one egress module for transmitting the incoming data to at least one egress port. The memory management unit is configured to receive data at a clock speed for the network device and write the data to the memory using a multiplied clock speed that is a multiple of the clock speed for the network device, read out the data from the memory at the multiplied clock speed and provide the data to the at least one egress module at the clock speed for the network device, where the multiplied clock speed is used to sample the clock speed for the network device to place domains of the multiplied clock speed and the clock speed for the network device in phase.

Abstract: A method includes receiving packets from a network with a plurality of packet-forwarding engines (PFEs) of a router, wherein the plurality of PFEs are interconnected by a switch fabric, determining an egress one of the PFEs for each of the packets, and forming fixed-sized fabric cells that share data associated with the packets that are destined for the same egress PFE while preventing packets destined for different egress PFEs to share any of the fabric cells. The fabric cells are transmitted through the switch fabric to communicate the packets to the egress PFEs.

Abstract: Method and system for hardware packet pacing using a direct memory access controller in a parallel computer which, in one aspect, keeps track of a total number of bytes put on the network as a result of a remote get operation, using a hardware token counter.

Abstract: A method of processing first and second record packets of real-time information includes computing for each packet a deadline interval and ordering processing of the packets according to the respective deadline intervals. A single-chip integrated circuit has a processor circuit and embedded electronic instructions forming an egress packet control establishing an egress scheduling list structure and operations in the processor circuit that extract a packet deadline intervals, place packets in the egress scheduling list according to deadline intervals; and embed a decoder that decodes the packets according to a priority depending to their deadline intervals.

Abstract: Data units received by a network device may be classified into traffic flow classes in which the determined traffic flow class for a data unit may be dynamically refined as the data unit is processed by the network device. A dispatch component of the network device may receive data units associated with traffic flow classes. Parallel processing engines of the network device may receive the data units from the dispatch component and may generate, for a least one of the data units, a plurality of dynamically refined indications of the traffic flow class to which the data unit belongs. Additionally, an ordering component of the network device may include a plurality of re-order queues, where the at least one data unit successively progresses through at least two of the re-order queues in an order defined by the plurality of dynamically refined indications of the traffic flow class.

Abstract: A network device receives a label-switched-path (LSP) labeled data packet, maps the LSP labeled data packet to an input queue, maps a data packet in the input queue to an output queue based on a received LSP label value and a received exp label value, and transmits the LSP labeled data packet from the output queue.

Abstract: Embodiments of the invention are directed to monitoring resources of a network processor to detect a condition of exhaustion in one or more of the resources over a predetermined time interval and to provide an indication of the condition. Some embodiments periodically sample various resources of a network processor and from the samples calculate utilization of the network processor's memory bus and core processor, and determine if an interworking FIFO packet queue error has occurred. Such information may help network operators and/or support engineers to quickly zero in on the root cause and take corrective actions for network failures which previously could have been attributed to many different causes and that would have required significant time and effort to troubleshoot.

Abstract: A system, method, and computer program product are provided for processing different content each stored in one of a plurality of queues. In use, a plurality of different content is identified for processing. Additionally, each of the different content is stored in one of a plurality of queues based on a classification thereof. Furthermore, the plurality of different content stored in the plurality of queues is processed.

Abstract: A system, method, and computer program product are provided for optimal packet flow in a multi-processor system on a chip. In operation, a credit is allocated for each of a plurality of agents coupled to a messaging network, the allocating including reserving one or more entries in a receive queue of at least one of the plurality of agents. Additionally, a first credit is decremented in response to a first agent sending a message to a second agent, the plurality of agents including the first and second agents. Furthermore, one of the first credit or a second credit is incremented in response to a signal from the second agent.