Apparatus and methods for reduction of transmission delay in a communication network
09641456 ยท 2017-05-02
Assignee
Inventors
- Yitshak Ohana (Jerusalem, IL)
- Philippe Klein (Jerusalem, IL)
- Avi Kliger (Ramat Gan, IL)
- Stephen Palm (Irvine, CA, US)
Cpc classification
International classification
Abstract
Apparatus and methods for reducing latency in coordinated networks are provided. The apparatus and methods relate to a protocol that may be referred to as the Persistent Reservation Request (p-RR), which may be viewed as a type of RR (reservation request). p-RR's may reduce latency, on average, to one MAP cycle or less. A p-RR may be used to facilitate Ethernet audiovisual bridging. Apparatus and methods of the invention may be used in connection with coaxial cable based networks that serve as a backbone for a managed network, which may interface with a package switched network.
Claims
1. A method comprising: receiving from a first node a persistent reservation request for allocation of network resources for transmission, by the first node via a first networking specification, of data frames not yet received by the first node from a second node via a second networking specification that differs from the first networking specification, wherein the persistent reservation request is for transmission of a first data frame in a first media access plan (MAP) cycle and for transmission of a second data frame in a second MAP cycle, each of said first data frame and said second data frame that has not yet been received by the first node, each of the first data frame and the second data frame being addressed via the second networking specification to a third node, the persistent reservation request having been generated responsive to a reservation message received by the first node from the second node via the second networking specification, characteristics of the persistent reservation request and the first and second data frames for transmission being indicated by the first networking specification, characteristics of the data frames not yet received by the first node indicated by the second networking specification, and the first node being in communication with a first network that corresponds to the first networking specification and a second network that corresponds to the second networking specification; and establishing a latency-reducing data flow comprising the first data frame and the second data frame, said establishing being based, as least in part, on the persistent reservation request, wherein the latency-reducing data flow is initiated prior to both the first data frame and the second data frame being received by the first node.
2. The method of claim 1 wherein the persistent reservation request comprises a traffic specification.
3. The method of claim 2 further comprising canceling the latency-reducing data flow based on information included in the traffic specification.
4. The method of claim 1 further comprising canceling the latency-reducing data flow based on execution of a link maintenance operation process.
5. The method of claim 1, wherein the establishing comprises allocating network resources during a portion of at least one medium access plan cycle, the portion having a duration that is less than a duration of the medium access plan cycle.
6. The method of claim 1, wherein the first networking specification comprises a non-Ethernet networking specification and the second networking specification comprises an Ethernet networking specification.
7. A device comprising: a receiver circuit in communication with a first network, the receiver circuit configured to receive a first reservation request from a first node in communication with the first network, wherein the first reservation request is for allocation of network resources for transmission, by a transmitter circuit of the first node via a first communication protocol of the first network, of at least a first and a second data frame not yet received by the first node from a second node via a second communication protocol of a second network, each of the first data frame and the second data frame being addressed via the second communication protocol to a third node, the first reservation request having been generated responsive to a reservation message received by the first node, characteristics of the first reservation request being indicated by the first communication protocol and characteristics of the first and second data frames not yet received being indicated by the second communication protocol; a processor circuit in communication with the receiver circuit, the processor circuit configured to establish a latency-reducing data flow based on the first reservation request; and the transmitter circuit configured to inform the first node of a grant of the first reservation request, the grant authorizing the first node to transmit the first data frame in a first media access plan (MAP) cycle and the second data frame in a second MAP cycle, wherein the transmitter circuit is configured to inform the first node of the grant of the first reservation request prior to the first node receiving the second data frame.
8. A method comprising: receiving, at a first node over a first network that corresponds to a first networking specification, a registration request, the registration request requesting receipt of multicast information by a second node from a third node, the registration request including a traffic specification; transmitting, using the first node, the registration request to the third node over a second network that corresponds to a second networking specification that differs from the first networking specification; receiving, at the first node and responsive to the registration request, a first reservation request from the third node over the second network; generating, using the first node and in response to receiving the first reservation request, a second reservation request for transmission over the first network, the second reservation request differing from the first reservation request; and transmitting, using the first node, the second reservation request to a network coordinator of the first network, the second reservation request conforming to the traffic specification, wherein: the first node transmits at least two frames of multicast data over the first network, each transmission occurring in a discrete media access plan (MAP) cycle prior to transmitting a third reservation request to the network coordinator of the first network.
9. The method of claim 8 further comprising, when the network coordinator generates medium access plan cycles and the second reservation request corresponds to one of the media access plan cycles, transmitting at least one additional reservation request, each of the at least one additional reservation request being transmitted in a respective subsequent media access plan cycle.
10. The method of claim 8, further comprising: receiving, by the first node, a grant from the network coordinator in response to the second reservation request, the grant not being provided to the third node.
11. A method comprising: transmitting a reservation request to a network coordinator for allocation of network resources on a non-packet-switched network segment during at least two medium access plan cycles for data that has not been received; receiving, subsequent to transmitting the reservation request, the data from a first packet-switched network segment using a packet-switched network ingress port, wherein the data is addressed to a node of a second packet-switched network segment via a communication protocol associated with the second packet-switched network segment; and transmitting the data over the non-packet-switched network segment to the second packet-switched network segment using a non-packet-switched network egress port.
12. The method of claim 11 further comprising, using the non-packet-switched network egress port, transmitting to the network coordinator a data flow request for data flow from the packet-switched network ingress port to the non-packet-switched network egress port.
13. The method of claim 11 further comprising: transmitting to the network coordinator the reservation request for the allocation of network resources for transmitting the data to the non-packet-switched network egress port; and transmitting at least two frames of the data before transmitting a second reservation request to the network coordinator.
14. The method of claim 11, further comprising, when the network coordinator generates medium access plan cycles: transmitting to the network coordinator the reservation request for the allocation of network resources for transmitting the data to the non-packet-switched network egress port, the reservation request corresponding to one of the generated medium access plan cycles; and transmitting at least one additional reservation request, each of the at least one additional reservation requests being transmitted in a respective subsequent medium access plan cycle.
15. The method of claim 14 wherein communication over the non-packet-switched network segment is performed in accordance with protocols defined by a specification relating to transmission of data over a shared coaxial medium.
16. The method of claim 11, further comprising receiving a first reservation request from the first packet-switched network segment, wherein transmitting the reservation request is in response to receiving the first reservation request.
17. The method of claim 11, wherein the data is received with characteristics of a first communication protocol associated with the first packet-switched network segment and the data is transmitted with characteristics of a second communication protocol associated with the non-packet-switched network segment.
18. A method comprising: transmitting, by a first node over a non-Ethernet network, a reservation request for requesting allocation of network resources for transmission, over the non-Ethernet network, of data frames not yet received by the first node over an Ethernet network, wherein the reservation request is for transmission of a first data frame in a first media access plan (MAP) cycle of the non-Ethernet network and for transmission of a second data frame in a second MAP cycle of the non-Ethernet network, each of said first data frame and said second data frame that has not yet been received by the first node from a second node over the Ethernet network at a time of the transmitting the reservation request, each of the first data frame and the second data frame being addressed via an Ethernet specification to a third node, the reservation request being transmitted in response to a reservation message received by the first node over the Ethernet network, the first node being in communication with the Ethernet network and the non-Ethernet network; and in response to the reservation request, receiving from a network controller instructions to transmit the first data frame in the first MAP cycle of the non-Ethernet network and the second data frame in the second MAP cycle of the non-Ethernet network.
19. The method of claim 18 wherein the reservation request comprises a traffic specification.
20. The method of claim 18 further comprising: transmitting the first data frame in the first MAP cycle and the second data frame in the second MAP cycle without transmitting a subsequent reservation request.
21. The method of claim 20, wherein the first data frame comprises a multicast data frame.
22. The method of claim 18, wherein the allocation of network resources is for a portion of at least one medium access plan cycle, the portion having a duration that is less than a duration of the medium access plan cycle.
23. A bridge device comprising: a processor circuit configured to establish a latency-reducing data flow on a first network based on a reservation request, the latency-reducing data flow comprising a first data frame and a second data frame received from a first node over a second network that differs from the first network in at least message passing characteristics; a transmitter circuit in communication with the first network, the second network, and the processor circuit, the transmitter circuit configured to transmit the reservation request in response to a reservation message received by the bridge device over the second network and prior to receipt of the first data frame and the second data frame by the bridge device over the second network, each of the first data frame and the second data frame being addressed to a second node via a network specification associated with the second network; and a receiver circuit configured to receive a grant of the reservation request from a network controller of the first network, the grant authorizing the bridge device to transmit the first data frame in a first media access plan (MAP) cycle of the first network and the second data frame in a second MAP cycle of the first network.
24. The bridge device of claim 23, wherein: the receiver circuit is further configured to receive the reservation message from the first node over the second network; and the transmitter circuit is configured to transmit the reservation request over the first network.
25. A device comprising: at least one processor circuit configured to: receive, over a first network, a registration request, the registration request requesting receipt of multicast information by a first node from a second node, the registration request including a traffic specification; transmit, over a second network that differs from the first network, the registration request to the second node; receive, over the second network, a first reservation request from the second node responsive to the registration request; and generate, responsive to the first reservation request received over the first network, a second reservation request for transmission over the second network; transmit the second reservation request to a network coordinator of the first network, the second reservation request conforming to the traffic specification; and transmit at least two frames of multicast data over one or more discrete media access plan (MAP) cycles of the first network prior to a transmission of a third reservation request to the network coordinator of the first network.
26. The device of claim 25, wherein the at least one processor circuit is configured to transmit the at least two frames over multiple discrete MAP cycles prior to the transmission of the third reservation request.
27. The device of claim 25, wherein the at least one processor circuit is configured to transmit the at least two frames in one discrete MAP cycle.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) The above and other features of the present invention, its nature and various advantages will be more apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings, and in which:
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
DETAILED DESCRIPTION OF EMBODIMENTS
(12) Apparatus and methods for reducing latency in coordinated networks are provided in accordance with the principles of the invention. The apparatus and methods relate to a protocol that may be referred to herein as the Persistent Reservation Request (p-RR), which may be viewed as a type of RR (reservation request). In some embodiments of the invention, p-RR can reduce latencies, on average, to one MAP cycle or less. In some embodiments of the invention, a persistent reservation request may be used to facilitate Ethernet audiovisual bridging. Apparatus and methods of the invention may be used in connection with networks that are described in one or more of MoCA Specifications v1.0, February 2006; MoCA Specifications v1.1 Extensions, May 2007; and IEEE802.1 AVB WG Specifications, all of which are incorporated by reference herein in their entireties.
(13) Methods in accordance with the principles of the invention may include a method for managing transmission of data over a shared media communication network. The method may include receiving from a node a first reservation request. The reservation request may request allocation of network resources for transmitting the data. The method may further include establishing a data flow based on the request, wherein the node transmits at least two data frames before a second reservation request is received. The flow may have a duration that is greater than the duration of one medium access plan cycle. The reservation request may include a traffic specification (TSPEC). Some embodiments of the invention may include canceling the flow based on information included in the traffic specification and/or canceling the flow based on the execution of a link maintenance operation (LMO) process. Some embodiments of the invention may include managing jitter by allocating network resources during only a portion of one or more MAP cycles.
(14) An exemplary device in accordance with the principles of the invention may be configured to manage transmission of data over a shared media communication network. The device may include a receiver module configured to receive a first reservation request from a node in communication with the network; a processor module configured to allocate network resources based on the request; and a transmitter module configured to inform the node about a grant of the request. The grant may authorize the node to transmit at least two data frames before the node transmits a second reservation request.
(15) Some methods in accordance with the principles of the invention may include providing Ethernet AV bridging using coordinated shared media networks. Those methods may include transmitting multicast information from a first node to a second node. The first node may reside in a packet mode network and the second node may reside in a packet-switched network. The method may include receiving at the first node a registration request from the second node. The registration request may request receipt of the multicast information and may include a TSPEC. The method may further include transmitting, via the first node, a reservation request to a network coordinator residing in the packet mode network. The reservation request may conform to the traffic specification.
(16) Some methods for Ethernet AV bridging in accordance with the principles of the invention may include using a packet mode network to transmit data from a first packet-switched network segment to a second packet-switched network segment. The methods may include receiving the data from the first segment using a packet mode network ingress port and transmitting the data to the second segment using a packet mode network egress port.
(17)
(18)
(19) The aforementioned MoCA 1.1 specification includes protocol pQoS (parameterized QoS), which supports parameterized QoS (Quality of Service) flow transactions over a network. In pQoS flow, a transmitting node sends a TSPEC admission request to a receiving node, as described in MoCA 1.1 section 4.2. After the request is accepted, a flow is created and data transmission between the transmitting and receiving nodes occurs in accordance with the protocol described above in connection with
(20) In some embodiments of the invention, after a pQoS flow is created, the transmitting node may use the flow's TSPEC to generate a p-RR for the flow and send the p-RR to the network coordinator (NC). Table 1 shows illustrative information that may be included in the p-RR
(21) TABLE-US-00001 TABLE 1 Illustrative p-RR Information p-RR Information Characteristics Leased time Copied from the flow's TSPEC. Indicates length of time period during which the p-RR applies. Packet duration Derived from the TSPEC nominal packet size and calculated for the connection current PHY profile (bit loading). Mean time Indicates the required mean time-between-transmission-grants to meet latency requirements (multiple transmissions may be allocated by the network controller to a flow within a single MAP cycle)
(22) In some embodiments, after the NC receives the p-RR, it can allocate to the transmitting node network resources for the flow based on the p-RR information, such as the leased time. In some embodiments, the p-RR may be allowed to persist (and is thus referred to as a persistent grant or PG) until a new p-RR for the same flow is received. When a node transmits exclusively audiovisual bridge traffic, and is operating under a persistent grant, there is no need to poll it frequently (every or almost every MAP CYCLE) as is usually done in MoCA1.0 networks. This may reduce overhead and improve network throughput.
(23) In response to the p-RR, the NC may allocate to the transmitting node a time-limited transmission opportunity. The time interval between consecutive transmission opportunities (viz., the transmission (Tx) service time), could be significantly smaller than the MAP cycle duration to accommodate the requirements of specific time sensitive traffic moving over the network. In some embodiments, a maximum Tx service time may be 100 s when the MAP cycle duration is 1 ms.
(24) If an LMO process (MoCA 1.0 section 3.7) causes a change in the connection profile associated with the transmitting and receiving nodes, the transmitting node may send an updated p-RR. If the data rate between the transmitting and receiving nodes changes, for example, as a result of an LMO process, the transmitting node may renew its p-RR. If the transmitting node fails to renew its p-RR, the NC may, after a predetermined number of MAP cycles, discontinue p-RR grants for the flow and send to the transmitting node a new p-RR opportunity. The NC may allocate a p-RR opportunity to the transmitting node after each LMO process between the transmitting and the receiving nodes. The allocation of a p-RR opportunity after each LMO process may be conditioned on the time remaining before the leased time expires.
(25)
(26) Packet switched network data streaming protocols (such as the stream reservation protocol (SRP) defined by IEEE standard 802.1Qat) often require registration by a listener node. The registration identifies to a talker node that the listener node desires to receive a data stream. After registration, such protocols require that the talker node transmit a reservation to reserve network resources along a path from talker to listener.
(27)
(28) From bridge 412, registration 450 is propagated to talker 418. It will be appreciated that inter-node communications regarding registration, notification, reservation and other information may require suitable modification of the communications, but for the sake of clarity, such modifications are ignored in
(29)
(30) When bridge 412 receives reservation message 470 from switch 416 (in packet switched network segment 406), bridge 412 may reject or accept reservation message 470. Bridge 412 may do so based on the availability of bridge 412 resources. Bridge 412 may query network coordinator 408 regarding the availability of resources in packet mode network segment 404. Network coordinator 408 may communicate reservation message 470 to bridge 410 along path 476. Bridge 410 may then propagate reservation message 470, via switch 414, to listener node 420.
(31) The P802.1Qat SRP is a one-way declarative protocol with reservation messages propagated from the talker towards listeners. It typically contains no backward propagated acknowledgement or status report messages. If needed, a talker node could leverage higher layer applications for getting feedback from listeners. A talker can also simply initiate the reservation message and then wait enough time before starting data transmission to ensure that the stream data can be served appropriately.
(32) A registration event may be initiated by a listener and passed by intermediate bridges, such as those of network segment 404. The talker node may be triggered by the registration event, which indicates that the listener node desires to establish a stream (data flow). Upon the registration event, the talker node may attempt to reserve the required resources in the local node and, for a shared media LAN, the talker may attempt to reserve resources on the LAN to which the talker node's egress port is attached. The talker may configure itself appropriately according to the local and LAN reservation message results. It then may record the result into the status information field of an updated reservation message and send the message toward the listener node.
(33) SRP assumes that an admission control policy (which governs the disposition of a request by a node that resides in one network segment for network resources that reside in a different network segment) is implemented by a network segment egress port (e.g., switch 416, shown in
(34) Reservation messages may be propagated over a subtree (e.g., a virtual LAN) by which the talker and listeners are connected. For each receiving bridge along the path from the talker to listeners, reservation messages can convey the result of resource reservation of the upstream bridges, and may trigger any necessary local or shared media LAN resource reservation operations.
(35) In certain embodiments of the invention, a talker node, such as 418, may refresh a reservation message on a regular basis. Each listener, such as 420, may keep a timer which will timeout the registration request if there is no corresponding reservation message received during the timer period.
(36) On receiving the first reservation message of a stream, some embodiments of the invention may allow a bridge to create a reservation record for the reservation message. The reservation record may contain information obtained from the reservation message. This information can include stream identifier, talker MAC address, traffic specification, upstream reservation status, reservation message inbound port and any other suitable information.
(37) If a bridge, such as 412 (shown in
(38) Because SRP is a one way declarative protocol, a failed reservation in an intermediate bridge will not influence any reservation that has been made upstream. Relevant listeners can receive reservation messages with negative reservation status. For example, a listener could choose to either withdraw the registration (e.g., by sending an update message corresponding to the registration), therefore releasing any unnecessary reserved resources, or keep the registration. By keeping the registration, the stream reservation can be made along the whole path when all necessary resources become available.
(39) If the received reservation message carries a negative reservation status, the bridge can preferably configure its forwarding resources appropriately, update the per-hop resource details information in the reservation message, and then transmit the reservation message out of each registered egress port.
(40) After a reservation record has been created, a bridge could later receive a reservation message that is inconsistent with the reservation record. While a discrepancy in the talker MAC address may be reported as an error and the reservation message rejected, a discrepancy in traffic specification, upstream reservation status, and reservation message inbound port may be accepted or rejected by the bridge based on predetermined policy.
(41)
(42)
(43)
(44)
(45) For the sake of clarity, the foregoing description, including specific examples of parameters or parameter values, is sometimes specific to certain protocols such as those identified with the name MoCA and/or Ethernet protocols. However, this is not intended to be limiting and the invention may be suitably generalized to other protocols and/or other packet protocols. The use of terms that may be specific to a particular protocol such as that identified by the name MoCA or Ethernet to describe a particular feature or embodiment is not intended to limit the scope of that feature or embodiment to that protocol specifically; instead the terms are used generally and are each intended to include parallel and similar terms defined under other protocols.
(46) It will be appreciated that software components of the present invention including programs and data may, if desired, be implemented in ROM (read only memory) form, including CD-ROMs, EPROMs and EEPROMs, or may be stored in any other suitable computer-readable medium such as but not limited to discs of various kinds, cards of various kinds and RAMs. Components described herein as software may, alternatively, be implemented wholly or partly in hardware, if desired, using conventional techniques.
(47) Thus, systems and methods for compensating for managing transmission of data over a shared media communication network and for facilitating Ethernet audiovisual bridging have been provided. Persons skilled in the art will appreciate that the present invention can be practiced using embodiments of the invention other than those described, which are presented for purposes of illustration rather than of limitation. The present invention is limited only by the claims which follow.