WO2009151847A1 - Network data transport multiplexer bus with global and local optimization of capacity allocation - Google Patents
Network data transport multiplexer bus with global and local optimization of capacity allocation Download PDFInfo
- Publication number
- WO2009151847A1 WO2009151847A1 PCT/US2009/043506 US2009043506W WO2009151847A1 WO 2009151847 A1 WO2009151847 A1 WO 2009151847A1 US 2009043506 W US2009043506 W US 2009043506W WO 2009151847 A1 WO2009151847 A1 WO 2009151847A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- network
- node
- bus
- source
- channels
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/52—Queue scheduling by attributing bandwidth to queues
- H04L47/525—Queue scheduling by attributing bandwidth to queues by redistribution of residual bandwidth
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/52—Queue scheduling by attributing bandwidth to queues
- H04L47/522—Dynamic queue service slot or variable bandwidth allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/625—Queue scheduling characterised by scheduling criteria for service slots or service orders
- H04L47/6255—Queue scheduling characterised by scheduling criteria for service slots or service orders queue load conditions, e.g. longest queue first
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/16—Time-division multiplex systems in which the time allocation to individual channels within a transmission cycle is variable, e.g. to accommodate varying complexity of signals, to vary number of channels transmitted
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/40—Bus networks
- H04L12/407—Bus networks with decentralised control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
Definitions
- the invention pertains generally to the field of communications network systems, and in particular to techniques for dynamically optimizing network capacity allocation based on data load variations across the network to maximize the network data throughput. [0004] Certain acronyms used in this specification are defined below:
- Ll ISO OSI model network protocol layer 1 i.e., the physical layer
- L2 ISO OSI model network protocol layer 2 i.e., the data link layer
- L3 ISO OSI model network protocol layer 3 i.e., the network layer
- Embodiments of the invention provide systems and methods for maximizing network data throughput by cyclically optimizing network capacity allocation according to relative quantities of data arrived, during a previous network capacity allocation cycle, at buffers at the network source nodes queuing data for future transport over the network to a network destination node associated with said buffers at the source nodes.
- a network throughput maximization system comprises a network transporting data from a set of source nodes to a destination node of the network. For each network capacity allocation cycle, counters at the source nodes monitor amounts of data written in buffers at their nodes queuing data to be transported to the destination node.
- An element at the destination node cyclically allocates the network data transport capacity among the set of source nodes, with the source nodes capable of reallocating and reassigning units of the network capacity, i.e. channels in the network, between themselves under rules defined for such local optimization, in order to achieve faster response of the network capacity allocation optimization process.
- This optimization of network data transport capacity allocation is performed, at least in part, according to relative volumes of data bytes written during a previous capacity allocation cycle into the source nodes buffers related to the network destination node.
- a network data transport capacity allocation optimization method involves monitoring at network source nodes the amounts of bytes of data packets that have arrived at the source nodes destined to a network destination node.
- the network destination node then cyclically globally optimizes the allocation of the network capacity among the network source nodes in view of the relative amounts of bytes received at all the source node buffers associated with the destination node during a previous network capacity allocation cycle, with the source nodes being able to reallocate and reassign channels within the network capacity between themselves, i.e., to override parts of the global network capacity allocation first done by the destination node according to their local optimization algorithms, in order to speed up the responsiveness of the network capacity allocation optimization process.
- a network data throughput maximization method involves globally optimizing network data transport capacity allocation by the destination node of a network, and locally optimizing same via readjusting the transport capacity allocation by the source nodes.
- This periodic method comprises a set of repeating sub-processes that include a process for receiving network data traffic at the destination node from a number of source nodes via a corresponding number of physical layer connections of variable capacities, a process for automatically adjusting the capacities of the source-node-specific connections by the destination node based at least in part on variations in data volumes associated with the connections from the source nodes to the destination node, and a process for readjusting the capacities of the physical layer connections by the source nodes.
- FIG. 1 shows an overview of a functional architecture of the network data throughput maximization system involving network buses that comprise a set of data transport channels, in accordance with an embodiment of the invention.
- FIG. 2 presents a row-based signal framing structure on a channel of a network bus of FIG. 1, comprising row overhead and payload fields, in accordance with an embodiment of the invention.
- FIG. 3 presents a more detailed view of the row overhead field of FIG. 2, in accordance with an embodiment of the invention.
- FIG. 4 presents in more detail a subfield of the row overhead field shown in FIG.
- FIG. 5 illustrates an architecture of a source node functionality in the network per
- FIG. 1 in accordance with an embodiment of the invention.
- FIG. 6 illustrates an architecture of a destination node functionality in the network per FIG. 1, in accordance with an embodiment of the invention.
- FIG. 7 illustrates an aspect of an embodiment of the invention where a source node in a network, such as the network shown in FIG. 1, reallocates its surplus units of network capacity to downstream source nodes.
- FIG. 8 illustrates an aspect of an embodiment of the invention where a source node in a network, such as the network shown in FIG. 1, reallocates to itself units of network capacity initially allocated to downstream source nodes, in order to match its demand of network capacity up to a level of fair division of network capacity among the source nodes.
- a box drawn with a dotted line indicates that the set of objects inside such a box form an object of higher abstraction level.
- FIG. 1 presents, in accordance with an embodiment of the invention, a multi- source-node digital communications bus 9, formed of its member channels 8, in a communications network system, which forwards and transports data traffic among nodes 4.
- data traffic or data refers to any form of digital data, including but not limited to digital video, digital voice and regular file transfer.
- bus 9(e) and channel 8(e) are able to transport data from the bus source nodes 11 to the bus destination node 10, which are the same as the source and destination nodes, respectively, of the member channels 8 of the bus 9.
- the buses 9 provide source node 11 to the destination node 10 Ll connections 6 whose capacities are cyclically optimized according to the data inflow volumes of the connections at their source nodes 11, by way of adjusting the number of bus channels 8 allocated for each source node specific connection 6 on a given bus 9 based on the variations in the relative amounts of bus capacity demanded by the source nodes of the bus.
- Various embodiments of the invented systems and methods can comprise any number of nodes 4, buses 9, etc. elements, as well as external interfaces to elements outside of the nodes 4.
- the bus 9(a) can have source nodes and their respective connections 6 from nodes other than those shown in FIG. 1.
- a channel 8 has a framed digital carrier, i.e., server signal, such as an SDH VC-3 aligned to an STM-N via an AU-3, and a source node 11 of the channel 8 is able to transmit data on the channel on a signal frame or a portion thereof, i.e., the source node is able to map data on at least a portion of the payload of the signal frame.
- server signal such as an SDH VC-3 aligned to an STM-N via an AU-3
- a source node 11 of the channel 8 is able to transmit data on the channel on a signal frame or a portion thereof, i.e., the source node is able to map data on at least a portion of the payload of the signal frame.
- a bus 9 can comprise any number of member channels, and in an embodiment of the invention each of the bus member channels 8 operate in alike manner described below with reference to FIGS. 2 through 6.
- the individual member channels 8 of the bus 9 are time-slot-interleaved within their bus envelope 9, in which case the bus 9 is said to use time- division-multiplexing (TDM) and its member channels 8 occupy their respective timeslots (TSs) within such TDM bus 9.
- TDM time- division-multiplexing
- TSs timeslots
- the VC-n based channels 8 are the M TSs interleaved in the VC-n-Mc carrier envelope 9, where the parameter n denotes an SDH path signal level such as 11, 12, 2, 3 or 4 or concatenation thereof, N any valid SDH physical signal level such as 0, 1, 4, 16, 64, 256 etc., and integer M the number of member VC-n TSs, i.e., channels 8 in the VC-n-Mc concatenation group 9.
- SDH STM-N i.e., STM-O/ 1/4/ 16/64/256 etc.
- the carrier signal frames on the bus 9 are data structures made of a number of rows, each comprising a number of columns, wherein each column on a row is a byte timeslot (BTS) of the signal. At least one of the columns on each row of the signal frames is an overhead column BTS, forming the row overhead field, while the other columns are for payload BTSs, forming the row payload field. Moreover, the consecutive rows within the signal frames are structured similarly with respect to the row overhead and payload columns.
- BTS byte timeslot
- FIG. 2 presents, in accordance with an embodiment of the invention, a sequence of signal rows 20 on a channel 8 of a bus 9 of FIG. 1, shown on a time axis 29, so that the signal data on the left end of the row sequence is transmitted and received the earliest.
- the signal rows in this context refer to any repeating, i.e., periodic blocks 20 of data fields, demarcated by row boundaries 27, on the channel signal.
- a periodic data block 20 can be a VC-3 row, which is repeated nine times per each VC-3 frame (and a VC-3 channel, mapped into e.g.
- SDH STM- N or SONET STS-N carrier signal comprises a repeating sequence of such VC-3 frames).
- a row 20 of FIG. 2 there exists several other type of embodiments for a row 20, e.g., SONET Virtual Tributaries.
- SONET Virtual Tributaries e.g., SONET Virtual Tributaries.
- VC-3 row is consistently used in this specification as an embodiment of the signal row 20 of FIG. 2, the discussion related to signal rows applies to any sort of repeating block or field of data within a digital signal.
- the exemplary row period 28 for this specification, i.e. VC-3 row period has the same nominal duration as that of an STM or STS row, i.e., one ninth of the 125ms frame period.
- the signal rows 20 of FIG. 2 have the following general characteristic:
- Each row 20 occupies a period of time referred to as row cycle 28, which has a constant nominal duration defined in applicable signal protocol standard, and which are separated in time from each other by row boundaries 27, a representative of which is marked in FIG. 2 between rows 20(a) and 20(b).
- a representative of a row cycle 28 is marked in FIG. 2 on the time axis 29 for the signal row cycle occupied by row 20(b);
- the rows 20 have a control information field 21, referred to as a row overhead;
- the rows 20 include a payload information field 22, referred to as a row payload, associated with them.
- the row overhead 21 is carried in the channel signal within the same time block 28 as is the payload 22 of the row 20, which time block is referred to as signal row cycle 28.
- the row control information 21 can be located anywhere in time within the row cycle 28, and does not need to be one continuous field as it is shown, for clarity, within the rows 20 in FIG. 2.
- the row overhead 21 used for the channel control signaling described herein is the VC-n path overhead (POH) column
- the row payload field 22 comprises the C-n payload columns of a VC-n row 20
- the row cycle 28 duration is nominally 125 ms divided by 9.
- the row control information fields 21 carry bitfields referred to as channel Access Control Tags (ACTs), which are used to identify one of the source nodes 11 of a channel of a bus 9, e.g., channel 8(e) of bus 9(e) in FIG. 1, to transmit data on that channel 8(e) on a row cycle 28 associated with each ACT.
- ACTs channel Access Control Tags
- the row cycle 28 associated with an ACT carried within the channel signal is defined in the applicable channel signal protocol specification, and a general principle is that the ACT is to precede in time the payload field 22 it is associated with.
- the ACT carried within the control information field 21 of a row 20 whose payload field 22 it is associated with identifies the active source node 11 for that row payload field 22 on the channel signal.
- the control information field 21 (a) of row 20(a) selects the active source node for mapping data on the payload field 22(a) of its row 20(a); the same way, 21(b) controls the transmission on 22(b), and so forth.
- the payload field 22(x) of a row 20(x) is defined as the payload field that begins during the same row cycle on which its associated control field 21(x) is located. Payload fields 22 can thus extend over a row cycle boundary 27 to the next row cycle.
- the row control information field 2 lean be carried in the first column of a VC-3 row, i.e., in the VC-3 POH column, and the row payload field 22 in the remaining columns in the C-3 payload of a VC-3 row.
- the payload field 22 can be mapped by an active channel source node 11 with any kind of client data of any protocol, as well as just raw data or idle bytes.
- the term byte is used in this specification to refer to a bit vector of any bit width (of at least one bit). While a common width of a byte is eight bits, in various embodiments a byte of data may comprise more or fewer then eight bits.
- FIG. 3 depicts, in accordance with an embodiment of the invention, a more detailed internal bitfield structure for an embodiment of the row overhead field 21 shown in FIG. 2.
- the format of the bitfields of the row overhead field shown as examples in FIGS. 3 and 4 are such that enable mapping the row overhead field 21 of a channel 8 to a single BTS, e.g., to the VC-3 POH column of a VC-3 framed channel as discussed above.
- the subfield 32 of the row control information field 21, comprising bits 32[O] and 32[1] in the embodiment presented in FIG.
- the ACT 32 identifies which one of the possible source nodes 11 for each channel 8(e), i.e., of nodes 4(a), 4(b), 4(c) and 4(d), is to map its data on the following row payload field 22 on that channel 8(e).
- the sub fields 31 of the row control information field 21 are for such a bus 9 source node 11, to which the destination node 10 of the bus is an end-of-bus source (EOB) node, to insert control information.
- EOB end-of-bus source
- node 4(a) is an EOB on the bus 9(e) to node 4(e).
- the same bit format for ACT is used in both subfields 31 and 32.
- That ACT format is such that it identifies the active source node, i.e., the node that is to transmit on the channel on the row payload 22 associated with the ACT 32, by expressing the number of source nodes 11 between the active node and the destination node, excluding both the active node and the destination node. That aforesaid value carried by ACT is referred to herein as the Active Node Identifier (ANI) for the row payload associated with the ACT.
- ANI Active Node Identifier
- SNIs Source Node Identifiers
- the channel access control involves: monitoring, by the individual source nodes of a channel 8, the ACT 32 field within the row overhead fields 21 of each signal row 20 on the channel; and transmitting data on the channel 8, on the row payload field 22 associated with the given ACT, by that channel source node 11 whose SNI matched the ANI carried by the ACT.
- SDH/SONET-based embodiment for signal framing such as VC-3
- VC-3 provides a straightforward way to extend the herein discussed dynamic, row cycle switched channel functionality into a multi-channel-bus via TDM byte-timeslot interleaving.
- a set of VC-3 channels in parallel to form a dynamic, row cycle switched data transport bus 9 for transport of data from the bus source nodes 11 to its destination node 10.
- Such a multi-channel bus 9 provides, besides a greater aggregate capacity than a single channel 8, also means for a more granular network bus capacity allocation among the source nodes 11 by enabling more than one, and even all, of the set of its source nodes 11 to transmit data on the same bus 9 to its destination node 10 concurrently, i.e., during the same row payload field 22, by way of allocating the member channels of the bus among multiple source nodes of the bus for a given row cycle.
- the source node 4(a) could be assigned the VC-3 channels #0, #1, #2 and #3, while node 4(b) could be assigned the channels #4, #5 and #6, 4(c) the channel #7 and #8, and 4(d) the rest, i.e., #9, #10 and #11.
- the allocation of the channels among the source nodes 4(a) through 4(d) of that 12xVC-3 bus 9(e) can be completely different, as defined by the ACTs 32 carried in the control fields 21 of individual VC-3 channels 8 on the subsequent VC-3 row 20.
- the set of bus channels assigned to an individual source node, such as node 4(c) of bus 9(e) in FIG. 1 are concatenated to continuously form a single, logically un-channelized, connection 6, e.g., 6(c) from node 4(c)) to the destination node of the bus 9.
- the EOB node e.g., node 4(a) on the channel 8(e) in FIG. 1
- this principle is met simply so that the EOB node transmits a continuous sequence of signal rows, with the payload fields 22 of exactly those rows 20 mapped with its client data that are identified as assigned to the EOB node by their associated ACTs 32 as these ACT bitfields are transmitted from the EOB source node downstream on the bus 9.
- the other source nodes 11, referred to as downstream source nodes, on the channel continuously keep track of the current row phase on the channel using a row BTS counter that is once per carrier signal frame period synchronized to the current frame phase on the channel signal as received from upstream channel, i.e., on the channel from the direction of its EOB node. Since the frames consist of similarly repeating rows 28, such as the nine 87-byte rows of a VC-3 frame, the frame phase synchronization also provides row phase synchronization for the source nodes of a channel.
- a downstream node along the channel 8 is able to locate and capture the channel-specific ACTs 32 on each row 20, as well to insert its own control information in its related subfields (e.g., 31 and 30) on each row transmitted on the downstream channel, and, incase the ACT matched with the SNI of the source node, map the row payload field 22 associated with the ACT with its client data to be transported to the destination node 10 of the channel 8.
- its related subfields e.g., 31 and 30
- the above described upstream-channel row-phase synchronized channel access process involves performing the SDH standard functions related to accessing the S3 data layer, i.e., VC-3 channel carried within an STM-N carrier signal, including STM-N frame boundary detection and AU-n pointer processing to locate the VC-3 frame and row boundaries 27 and VC-3 POH 21 and C- 3 row payload 22 BTSs.
- an efficient implementation for a multi-channel bus 9 formed of the above described BTS-synchronously accessed, parallel multi-source-node channels 8, is achieved so that all member channels of the bus have the same common frame and row phase when received by any of the downstream source nodes 11 or the bus destination node 10, so that the row cycles 28, and thus row boundaries 27, coincide throughout all channels 8 of the bus 9.
- this principle is upheld via arranging all the member VC-3 channels 8 of the bus 9 to be member VC-3 TSs of a single VC-3-Mc concatenation group, which is aligned to its carrier STM-N signal via a single AU-3-Mc pointer, thus ensuring that all member VC-3s of the VC-3-Mc bus stay in the same row BTS phase (based on that the row BTS phase of a VC-3 is defined by the frame phase of its STM- N server signal and the offset of the AU pointer with which the VC-3 is aligned to its STM- N).
- the channels 8 are VC-3 TSs within a single VC-3-Mc bus 9 transmitted by the EOB node of the bus comprising M (any integer greater than zero) VC-3 TS channels in total.
- the AU-3-Mc (including its pointer bytes, which are regenerated at each node according to standard SDH pointer processing) flow through all the way to the destination node of the bus, so that the destination node will continuously receive a regular SDH VC-3-Mc signal on the bus with a continuous AU-3-Mc pointer from frame to frame, regardless of which source node mapped data on which VC-3 TS channel 8 of the VC-3-Mc bus 9 on any given row payload field 22.
- each source node connection associated receiver instance at the bus destination node 10 knows which VC-3 TSs on the VC-3-Mc bus 9 belongs to its associated connection 6 on each row cycle on the bus.
- Such a source node connection associated receiver instance provides a SDH-to-packet demapper function in case of packet-based client traffic carried on its associated connection 6.
- Bitfield 31 in FIG. 2, comprising in this embodiment of bits 31 [0] and 31 [1], is used for carrying an ACT for a channel 8 on which the node 4 generating the ACT is the destination node 10.
- the destination node 4(e) of a channel 8(e) inserts the ACTs controlling access for the channel 8(e) on the bitfields 31 of the overhead fields 21 on another channel 8 on the bus 9(a), which delivers the ACTs to the EOB node 4(a) of the bus 9(e) channel 8(e).
- FIG. 4 presents the internal structure of subf ⁇ eld 30 of the channel signal row control field 21 shown in FIG. 3, in accordance with an embodiment of the invention.
- bitfield 30 is used to carry capacity request information, in units of VC-3 TSs 8 for a one VC-3 row cycle 28, by the source node inserting it for the bus 9 on which the bitfield 30 travels on.
- a bus source node 1 l(x) sets its associated bits 40(x) of the bitfields 30 on as many of the M VC-3 TS-channels 8 on the bus as is its current Bus Capacity Request (BCR) in units of VC-3 TSs for its connection 6 to the bus destination node 10.
- BCR Bus Capacity Request
- An embodiment of a bus source node 11 computes the magnitude of its BCR figure, i.e., its requested number of channel units on the bus, based on its amount of data bytes written, during the preceding row cycle 28, into buffers storing data for future delivery on the bus 9 toward its destination node 10.
- bitfield format of row control information field 21 described herein in reference to FIGS. 3 and 4, comprising eighth bits, in an embodiment of the invention can be carried in the POH column BTS of a VC-3 frame structured channel 8.
- Such an embodiment of the invention described herein does not use any of the Ll (SDH) signal payload capacity for control signaling, but instead allows using all the Ll payload capacity for carrying actual client data, e.g., L2 packets.
- Ll SDH
- FIG. 5 presents a block diagram of a source node 11 functionality, including the Bus Capacity Requests (BCR) generation, in a network per FIG. 1, in accordance with an embodiment of the invention.
- BCR Bus Capacity Requests
- B data bytes
- connection source buffer byte write counter logic 51 gets its row cycle boundary notification 27 from a row BTS counter 55 that operates in the bus 9 server signal frame and row phase, causing the amount of data bytes written for the associated connection source buffers to be sampled and recorded for each successive bus capacity allocation cycle.
- the BCR 57 of a source node i.e., the number M of VC-3 TSs the source node requests for its connection 6 on the bus for the next VC-3 row cycle, can be computed using the formula:
- BCR MIN ⁇ M, MAX ⁇ 1, INT(B/64) ⁇ ⁇ ;
- bus 9 capacity allocation unit is a VC-3 row, which provides 86 payload bytes
- a more accurate BCR generation algorithm is such where the byte inflow B associated with a connection 6 is divided by value 86.
- the combined data inflow figures for each of the buffers 52 which buffer data bytes queued for future transport to the destination node associated with the buffers and from which the bus mapper 53 reads 59 data to its associated bus 9, are used in forming the value of the data byte inflow B 56, based on which the value of the BCR 57 for that bus is computed by the BCR generator 54.
- the data bytes written 58 into the connection source buffers 52 can be received by the source node 11 from different interfaces of it, including from an access interface 7 over which the source node receives data from external systems, e.g., from an external L2 or L3 node, or from another bus 9.
- Embodiments of the invention can use also additional factors, besides the volume of the data byte inflow (B) during the previous bus capacity allocation cycle per above, in forming the value of their BCRs.
- additional factors can include, for example, the amount of data bytes queued at the source node buffers from which the bus mapper 53 maps data to its associated bus 9, as well as the priority levels of the connection data inflow or the buffered data.
- FIG. 6 presents a block diagram of a destination node functionality, in accordance with an embodiment of the invention.
- the bus 9 is received by a module referred to as the Bus Access Controller (BAC) 61.
- BAC Bus Access Controller
- the BAC periodically, for each new row cycle 28 on its bus 9, performs a bus channel 8 allocation optimization among its source nodes 11, assigning the channels of the bus between the source node specific connections 6 for a future bus capacity allocation cycle, based on the most recent set of BCRs it has received on the bus 9 from its source nodes on the source node associated bits 40 of sub fields 30 (FIGS. 3 and 4).
- the network capacity allocation optimization method of the invention is employed in order to continuously to maximize the rate of client data delivered by a network bus 9, and the capacity optimization algorithms performed by the destination node, in view of the BCRs from all the source nodes in parallel, is referred to herein as global bus capacity allocation optimization.
- This allocation and assignment of bus channels 8 among its source nodes 11 produces a set of bus channel 8 specific ACTs, each of which carries an ANI 32 identifying one of the bus source nodes 11 to map data on its associated row payload field 22 on the bus channel 8 that a given ACT travels on.
- the BAC 61 at the destination node 10(e) of the channel 9(e) sends the set of ACTs (using one of its local bus mappers 53) on subfield 31 of row overhead fields 21 on the bus 9(a) (on which that host node 4(e) of the BAC is a source node 11) to the bus 9(a) destination node 4(a), which is the bus 9(e) EOB node.
- That EOB node 4(a) of bus 9(e) will then loop the set of ACTs via subfields 32 back on the bus 9(e). All source nodes 4(a) through 4(d) process and apply the set of ACTs on the bus 9(e) (as described herein in references to FIGS. 3, 5, 6, 7 and 8), i.e., transmit data on the payload fields 22 associated with such ACTs 32 that carry an ANI that match their respective SNIs as the ACT bitfields are transmitted downstream on the bus from each source node.
- the BAC 61 performing capacity allocation for a bus 9 to which its host node 10 is the destination, in an embodiment of the invention per FIG.
- the packet demappers recover from the VC-n frames their respective L2 packet sequences 69 for further processing by downstream logic 63.
- the channel 8 specific ACTs 32 on the row control fields 21 of each row 20 on the bus 9 identify to which source node specific connection 6 their associated row payload field 21 belongs to.
- the BAC 61 at the bus destination node 10 thus knows which VC-3 TS channels 8 on the VC-3-Mc bus 9 belongs to which source node 11 specific connection 6 on each new payload field 22 on the bus 9. Accordingly, the BAC 61 enables the correct source node specific packet demapper 62 to receive data bytes on the bus 9 on each bus byte timeslot. That way, each of the demappers 62(a), 62(b), 62(c) and 64(d) per FIG.
- FIG. 6 receives a data byte sequence from the bus 9 that form the original data packet sequence 69(a), 69(b), 69(c), or 69(d) sent by its related source node 4(a), 4(b), 4(c) or 4(d), respectively, on their corresponding connection 6(a), 6(b), 6(c) or 6(d) on the bus 9(e) to its destination node 4(e).
- a packet multiplexer 63 that multiplexes data packets from the source node specific packet streams 69(a) through 69(d) into a combined packet stream 8 that can be sent over an external network interface for instance to another packet-switching or processing network element, or to another node 4 based on this specification.
- FIG. 7 illustrates an aspect of the invention, using the network per FIG. 1 as an example, where logic at a bus mapper 53 at a source node 4(a) through 4(c) reallocates and reassigns its local surplus units of network capacity to its downstream source nodes 4(b) through 4(d) along the bus 9(e).
- FIGS. 7.A and 7.B respectively, present a possible scenario of binary values of ACTs 32 before and after reallocation of a channel 8 from one source node 11 to another one downstream from it along a bus 9. Note that since local optimization algorithms performed by the source nodes 11, described herein in reference to FIGS 7 and 8, result both in reallocation of channels (i.e.
- a source node 11 such that has one or more source nodes between itself and the destination node along the bus 9 under study, reassigns from itself a maximum number (if any) of the channel 8 resources that were marked as assigned to it to its next downstream source node along the bus 9 such that after said reassignment the remaining number of channels assigned to the source node performing the reassignment still meets the amount of channel resources demanded by the most recent BCR figure of that node for the given bus under study.
- these local surplus channels being reassigned are the channels with the highest channel order numbers among the channels assigned to the node performing the reassignment by the ACTs received by said node.
- the node carries out such reassignment of channels by overwriting the ACT bits 32 of such surplus channels with the ACT field value that is one less than its own SNI value.
- FIG. 7 presents an example, situated for the network of FIG. 1, of a case of the source node 4(c) reassigning one of its originally assigned channels on the bus 9(e) to its next downstream source node 4(d), by rewriting the value in the received ACT bit field 32 of such a channel, i.e., overwriting decimal value 1 (shown in FIG. 7.A as binary 01) with a new value decimal 0 (shown in FIG. 7.B as binary 00) that is one less than its own SNI, which ACT value activates its next downstream node that in the case for source node 4(c) is the node 4(d).
- FIGS. 7 and 8 depict the case of using big-endian encoding for ACT value transmission.
- the node 4(c) receives six of the twelve channels on bus 9(e), specifically channels with order numbers of 3 through 8, but its most recent BCR figure for that bus demanded only four channels, the node 4(c) rewrites the ACT bit fields 32 for the channels numbers 7 and 8 with decimal value 0 in the signal that it transmits downstream on the bus 9(e), and accordingly does not map its data on the payload fields associated with these overwritten ACTs. Consequently, the node 4(d) receives the ACTs of these reassigned channel with ACT values that match its SNI i.e.
- the destination node despite any reassignment of local surplus channels among the source nodes along the bus, thus also gets the valid source-node to channel resource mapping info via ACTs that it receives on the bus.
- FIG. 8 illustrates an aspect of the invention, using the network of FIG. 1 as an example, where logic at a bus mapper 53 at a source node reassigns to itself units of network capacity initially assigned to its downstream source nodes to match its demand up to its fair share of the bus capacity, in a case when the source node had been allocated less that its fair share of bus capacity while its most recent BCR to the destination node demanded more channels on the bus that it had been allocated.
- FIGS. 8.A and 8.B respectively, present a possible scenario of binary values of ACTs 32 before and after reallocation of a channel 8 by a source node 11 to itself from another one source node 11 downstream from it along a bus 9.
- the fair share of bus 9 capacity for each of its source node 11 is equal to the lesser of i) total number of bus channel resources divided by the number of source nodes on the bus and ii) the data mapping capacity of a given source node on the bus, in units of bus transport channel 8 resources.
- the fair share of bus capacity can be individually configured for each source node along a given bus; however the sum of the source node specific fair shares of capacity along any given bus should match the total amount of channels on the bus.
- a source node 11 such that has at least one source node between itself and the destination node along the bus 9 under study, reassigns to itself a minimum sufficient number of channels (if any) that that meets the lesser of its latest BCR or its fair share of channels on the bus, however within the limits of the number of channels to be reassigned i.e. channels assigned to source nodes downstream from it on the bus.
- these local deficit channels being reassigned are the channels with the lowest channel order numbers among the channels that were assigned to the downstream node(s) before the reassignment discussed herein.
- the node carries out such reassignment of channels by overwriting the ACT bits 32 of the channels being reassigned with the ACT field value equal to its own SNI value.
- FIG. 8 presents an example, situated for the network of FIG.
- decimal value 2 shown in FIG. 8.A as binary 10
- decimal 3 shown in FIG. 8.B as binary 11
- the source node 4(a) accordingly also maps data on the channel payload field 22 that it reassigned to itself, along with other channels assigned to its via the ACTs that it received for that bus cycle, and correspondingly, the node 4(b) knows from the rewritten ACT for the reassigned channel to not map its data on its associated payload field 22 but to simply pass its through down the bus.
- the node 4(a) rewrites the ACT bit field 32 for the channel number 2 with decimal value 3 (equal to its SNI) in the signal that it transmits downstream on the bus 9(e).
- the node 4(b) receives the ACT of this reassigned channel with ACT value that does not match its SNI (of decimal value 2), causing the node 4(b) to treat the channel as being already used by one of its upstream nodes on the bus.
- the bus destination nodes 10 monitor the ACT 32 values received on the buses for determining which channels make up the connection 6 from which source node, regardless of any surplus and/or deficit based reassignment of channels among the source nodes 11 along the buses, the destinations of the buses thereby automatically get correct source-node to channel resource 8 mapping info via the ACT signaling received on the bus 9 for every bus row cycle 28.
- the source nodes use the BCR value that they use for encoding the bitfield 30 transmissions on the bus as the latest BCR figure for determining whether they have local surplus or deficit of channels allocated on the bus, and for a basis for consequent reassignments per FIGS. 7 and 8 and related descriptions.
- the source nodes also rewrite the bitfields 32 in connection with such reassignments in similar manner as they transmit their BCR figures encoded in bitfields 30.
- Bus capacity allocation and channel 8 realignment processes by the source nodes 11 along a bus 9 according to the principles above in references to FIGS. 7 and 8 achieve local optimization of bus capacity allocation, as techniques additional to the global i.e.
- all of the nodes 4 may include both source and destination node functionalities.
- any one of the discussed functions can be implemented either by their respective discrete logical modules (e.g. as illustrated in the examples of FIGS. 5 and 6), by a combination of a number of functional modules, or as embedded with other functions.
- the local optimization i.e.
- capacity allocation readjustments (functions discussed in connection with FIGS. 7 and 8) by bus source nodes could be performed at each source node by a logical module separate from the module performing the bus 9 data mapping, and in some embodiments, such capacity reallocation module could furthermore be combined with a general control module performing the global capacity allocation optimizations for other buses.
- functions related to the herein presented module BAC can be separated into various other modules, and those logical functions can be implemented for instance combined with the modules performing the demapping of data (illustrated in FIG. 6 as elements 62) from the buses 9.
- elements performing the functions presented herein in connection with FIGS. 1 through 8 can be also named differently, while still utilizing the invented systems and methods.
- Embodiments of the present invention include an efficient communications network architecture based on physical layer i.e. Ll connections whose capacities are automatically and continuously optimized according to their input data flow variations. [0075] Per the discussion in the foregoing regarding the drawings, embodiments of the invention perform traff ⁇ c-load-adaptive optimization of communications network capacity allocation, thereby maximizing data throughput of a communications network based on real-time data traffic load variations across the network, such as the example network system of FIG. 1.
- Such a network system comprises a data transport layer, called data plane, implemented with buses 9, and a control plane, implementation of which involves in an embodiment of the invention a logic module called Bus Access Controller (BAC) that is located at the destination node 10 of each bus 9 and that performs the global optimization phase of dynamic channel 8 allocation algorithms for buses 9.
- BAC Bus Access Controller
- One embodiment of the invention provides a byte-timeslot-synchronous, non- overhead-adding and data-loss-free data transport channel 8 based on a novel channel-access- control signaling mechanism enabling dynamic signal row (e.g., VC-3 row) cycle synchronous switching.
- a novel channel-access- control signaling mechanism enabling dynamic signal row (e.g., VC-3 row) cycle synchronous switching.
- multiple, potentially geographically distant and independently clocked source nodes 11 can map data on the same communication channel 8 so that a different source node can map data on the channel on every new row cycle 28 on the channel.
- switching between the old and new active source nodes on the bus takes place on a definite column on each row cycle 28 on the channel.
- the dynamic row cycle switched multi-source- node channel 8 operates so that the furthermost source node on the channel, called the end- of-bus source (EOB) node, continuously sends server signal frames, comprising repeating sequences of rows, on the bus with a pointer identifying the BTS location of the first byte of each channel frame sent.
- the EOB source signals, via the overhead field 21 of each row 20 of the signal frames, to which one, potentially itself, of the source nodes of the channel the next row payload field 22 on the channel is assigned to.
- a source node of the channel maps its data on the channel on exactly those row payload fields which are marked as assigned to it by the value of an active source node identifier carried in the preceding row overhead field as these overhead fields are transmitted from each given source node downstream on the bus.
- the row cycle herein can be any repeating block 20 of data on the channel signal that comprises the row overhead 21 and row payload 22 BTSs. For instance, a VC-3 row of BTSs fills that definition in case of a VC-3 framed channel 8.
- the BAC 61 adds together the Bus Capacity Request (BCR) bits 40 that are set in their active values in the source node 1 l(x) associated bits field 40(x) within the row control information fields 21 on each channel 8 of the bus 9, to gather the BCR figures, an integer number of requested channels, for each individual bus source node.
- BCR Bus Capacity Request
- the BAC then runs its global optimization algorithms to allocate and assign the bus channels 8 among the set of bus source nodes 11 once every bus row cycle 28 based on the most resent BCRs received from its bus source nodes, in one embodiment of the invention, according to the following algorithm: 1.) BAC allocates the member channels 8 of the bus 9 among the individual source node specific connections 6 so that it assigns channels, one channel at a time, per each source node whose BCR figure exceeds the number of channels so far assigned to it, until either all channels of the bus are allocated, or until the BCR of each source node is fulfilled, whichever comes first.
- the BAC initially assigns any remaining unallocated channels to the EOB source node of the bus, allowing the EOB source to reassign its local surplus channel resources to its next downstream source node, and so on by the source nodes down the bus, thus enabling the source nodes to perform their local optimization of the bus capacity allocation during the bus capacity allocation cycle on the bus, according to FIGS. 7 and 8 and related descriptions.
- This method of globally optimizing network bus capacity allocation produces a channel order number indexed table containing a set of Active Node Identifiers (ANIs) identifying the active source node per each bus channel 8 for a new row cycle 28 on the bus 9.
- ANIs Active Node Identifiers
- the bus source nodes 11 may perform local optimization of bus 9 capacity allocation, by readjusting the initial, global bus channel allocation by the destination node 10, based on their local channel surplus or deficit amounts, in order to achieve faster responsiveness of the bus capacity allocation optimization process in particular for cases of longer transmission distance based delays in the source-destination-source node control loop.
- BAC 61 For every new row cycle 28, BAC 61 distributes the channel 8 specific ANIs, i.e., bus Access Control Tags (ACTs), through its local bus mapper 53 accessing another bus 9, to the furthest most source node, i.e., EOB node of the bus to which the local node 4 of the BAC in question is the destination, using bus row control overhead field 31 based signaling. That EOB source node of the bus 9 under study then loops back the ACTs via bus row control overhead field 32, thereby signaling the active source node identification per each bus channel 8 for the bus source and destination nodes for each new row cycle 28 on the bus 9.
- ACTs bus Access Control Tags
- bus row overhead 21 subfields 30, 31 and 32 are described in detail in the foregoing regarding the drawings, in particular FIGS. 3 through 8. Reference material
Abstract
Description
Claims
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB1100501.4A GB2473174B (en) | 2008-06-12 | 2009-05-11 | Network data transport multiplexer bus with global and local optimization of capacity allocation |
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US6090508P | 2008-06-12 | 2008-06-12 | |
US61/060,905 | 2008-06-12 | ||
US7510808P | 2008-06-24 | 2008-06-24 | |
US61/075,108 | 2008-06-24 | ||
US9672608P | 2008-09-12 | 2008-09-12 | |
US61/096,726 | 2008-09-12 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2009151847A1 true WO2009151847A1 (en) | 2009-12-17 |
Family
ID=41414666
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2009/043506 WO2009151847A1 (en) | 2008-06-12 | 2009-05-11 | Network data transport multiplexer bus with global and local optimization of capacity allocation |
Country Status (3)
Country | Link |
---|---|
US (1) | US8804760B2 (en) |
GB (1) | GB2473174B (en) |
WO (1) | WO2009151847A1 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8842988B2 (en) | 2011-06-20 | 2014-09-23 | Plexxi Inc. | Optical junction nodes for use in data center networks |
US9204207B2 (en) | 2011-11-01 | 2015-12-01 | Plexxi Inc. | Hierarchy of control in a data center network |
US9288555B2 (en) | 2011-11-01 | 2016-03-15 | Plexxi Inc. | Data center network architecture |
US9301026B2 (en) | 2011-11-01 | 2016-03-29 | Plexxi Inc. | Affinity modeling in a data center network |
US9337931B2 (en) | 2011-11-01 | 2016-05-10 | Plexxi Inc. | Control and provisioning in a data center network with at least one central controller |
US9450815B2 (en) | 2013-07-11 | 2016-09-20 | Plexxi Inc. | Network node connection configuration |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
ES2652117T3 (en) * | 2013-07-04 | 2018-01-31 | Omicron Electronics Gmbh | Data transmission through a memory of a communication device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6097733A (en) * | 1997-06-13 | 2000-08-01 | Nortel Networks Corporation | System and associated method of operation for managing bandwidth in a wireless communication system supporting multimedia communications |
US6574222B1 (en) * | 1998-02-20 | 2003-06-03 | Sprint Communications Company, L.P. | System and method for allocating bandwidth for a call |
US7333511B2 (en) * | 2002-08-29 | 2008-02-19 | Optimum Communications Services, Inc. | Dynamically channelizable packet transport network |
US7349414B2 (en) * | 2001-08-24 | 2008-03-25 | Optimum Communications Services, Inc. | System and method for maximizing the traffic delivery capacity of packet transport networks via real-time traffic pattern based optimization of transport capacity allocation |
Family Cites Families (42)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5315596A (en) * | 1992-04-17 | 1994-05-24 | Canadian Institute For Broadband & Information Network Technologies, Inc. (Cibint) | Digital multiplexer with logically allocatable channels and bit rates |
US5506712A (en) * | 1993-07-14 | 1996-04-09 | Nippon Telegraph And Telephone Corporation | Photonic frequency routing type time division highway switch |
US5526344A (en) * | 1994-04-15 | 1996-06-11 | Dsc Communications Corporation | Multi-service switch for a telecommunications network |
US5461622A (en) * | 1994-06-14 | 1995-10-24 | Bell Communications Research, Inc. | Method and apparatus for using SONET overheat to align multiple inverse multiplexed data streams |
US5646758A (en) * | 1994-08-12 | 1997-07-08 | Nippon Telegraph And Telephone Corporation | Optical time compression multiplexing transmission system |
US5719858A (en) * | 1995-07-31 | 1998-02-17 | Paradyne Corporation | Time-division multiple-access method for packet transmission on shared synchronous serial buses |
GB9704587D0 (en) * | 1997-03-05 | 1997-04-23 | Fujitsu Ltd | Wavelength-division multiplexing in passive optical networks |
JP2000134246A (en) * | 1998-10-26 | 2000-05-12 | Fujitsu Ltd | Transmitter |
JP3705942B2 (en) * | 1998-10-30 | 2005-10-12 | 富士通株式会社 | Cross-connect switch |
US6782007B1 (en) * | 1999-01-26 | 2004-08-24 | Samsung Electronics Co., Ltd. | TDM bus synchronization circuit and protocol and method of operation |
JP2000269911A (en) * | 1999-03-17 | 2000-09-29 | Fujitsu Ltd | Transmission terminal equipment and network system |
US7110358B1 (en) * | 1999-05-14 | 2006-09-19 | Pmc-Sierra, Inc. | Method and apparatus for managing data traffic between a high capacity source and multiple destinations |
JP3539551B2 (en) * | 1999-11-16 | 2004-07-07 | 日本電気株式会社 | Frame and transmission device for accommodating heterogeneous data traffic on common carrier |
US7031263B1 (en) * | 2000-02-08 | 2006-04-18 | Cisco Technology, Inc. | Method and apparatus for network management system |
US20020049608A1 (en) * | 2000-03-03 | 2002-04-25 | Hartsell Neal D. | Systems and methods for providing differentiated business services in information management environments |
JP2001292164A (en) * | 2000-04-06 | 2001-10-19 | Nec Corp | Packet switch and its switching method |
US6975649B1 (en) * | 2000-05-25 | 2005-12-13 | Nortel Networks Limited | Hyper-concatenation across independent pointer processors |
US7693976B2 (en) * | 2000-07-11 | 2010-04-06 | Ciena Corporation | Granular management of network resources |
CA2354446A1 (en) * | 2000-11-22 | 2002-05-22 | Blaine Hobson | An optical switch and method of switching optical signals |
US7193968B1 (en) * | 2001-02-08 | 2007-03-20 | Cisco Technology, Inc. | Sample netflow for network traffic data collection |
US6973229B1 (en) * | 2001-02-28 | 2005-12-06 | Lambda Opticalsystems Corporation | Node architecture for modularized and reconfigurable optical networks, and methods and apparatus therefor |
US6807186B2 (en) * | 2001-04-27 | 2004-10-19 | Lsi Logic Corporation | Architectures for a single-stage grooming switch |
US7130276B2 (en) * | 2001-05-31 | 2006-10-31 | Turin Networks | Hybrid time division multiplexing and data transport |
US6883084B1 (en) * | 2001-07-25 | 2005-04-19 | University Of New Mexico | Reconfigurable data path processor |
US9048965B2 (en) * | 2001-08-24 | 2015-06-02 | Mark Henrik Sandstrom | Input-controllable dynamic cross-connect |
US20030074430A1 (en) * | 2001-10-05 | 2003-04-17 | Gieseke Eric James | Object oriented provisioning server object model |
US7319666B2 (en) * | 2002-02-06 | 2008-01-15 | Thomson Licensing | Method and apparatus for concatenating and piggybacking data packets |
US7254138B2 (en) * | 2002-02-11 | 2007-08-07 | Optimum Communications Services, Inc. | Transparent, look-up-free packet forwarding method for optimizing global network throughput based on real-time route status |
US7072361B1 (en) * | 2002-03-08 | 2006-07-04 | Applied Micro Circuits Corporation | System and method for the transport of backwards information between simplex devices |
US7558260B2 (en) * | 2003-03-07 | 2009-07-07 | Optimum Communication Services, Inc. | Byte-timeslot-synchronous, dynamically switched multi-source-node data transport bus system |
US20070083628A1 (en) * | 2005-10-11 | 2007-04-12 | Sandstrom Mark H | Automated, transparent and secure system and method for remotely managing network elements |
US7562393B2 (en) * | 2002-10-21 | 2009-07-14 | Alcatel-Lucent Usa Inc. | Mobility access gateway |
WO2004051897A1 (en) * | 2002-11-29 | 2004-06-17 | Fujitsu Limited | Packet transmission system and end office unit |
WO2005081672A2 (en) * | 2004-01-30 | 2005-09-09 | International Business Machines Corporation | Componentized automatic provisioning and management of computing environments for computing utilities |
US8458467B2 (en) * | 2005-06-21 | 2013-06-04 | Cisco Technology, Inc. | Method and apparatus for adaptive application message payload content transformation in a network infrastructure element |
US7724732B2 (en) * | 2005-03-04 | 2010-05-25 | Cisco Technology, Inc. | Secure multipoint internet protocol virtual private networks |
ATE389282T1 (en) * | 2005-04-19 | 2008-03-15 | Alcatel Lucent | METHOD FOR OPERATING A PACKET-BASED DATA NETWORK |
US20080037553A1 (en) * | 2005-12-22 | 2008-02-14 | Bellsouth Intellectual Property Corporation | Systems and methods for allocating bandwidth to ports in a computer network |
US8122174B2 (en) * | 2006-03-31 | 2012-02-21 | Research In Motion Limited | System and method for provisioning a remote resource for an electronic device |
US8199642B2 (en) * | 2006-09-14 | 2012-06-12 | Cisco Technology, Inc. | Dynamically and efficiently forming hierarchical tunnels |
US8315171B2 (en) * | 2006-10-31 | 2012-11-20 | Oracle America, Inc. | Adaptive management of computing resources |
US7986713B2 (en) * | 2006-12-09 | 2011-07-26 | Mark Henrik Sandstrom | Data byte load based network byte-timeslot allocation |
-
2009
- 2009-01-30 US US12/363,667 patent/US8804760B2/en active Active
- 2009-05-11 WO PCT/US2009/043506 patent/WO2009151847A1/en active Application Filing
- 2009-05-11 GB GB1100501.4A patent/GB2473174B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6097733A (en) * | 1997-06-13 | 2000-08-01 | Nortel Networks Corporation | System and associated method of operation for managing bandwidth in a wireless communication system supporting multimedia communications |
US6574222B1 (en) * | 1998-02-20 | 2003-06-03 | Sprint Communications Company, L.P. | System and method for allocating bandwidth for a call |
US7349414B2 (en) * | 2001-08-24 | 2008-03-25 | Optimum Communications Services, Inc. | System and method for maximizing the traffic delivery capacity of packet transport networks via real-time traffic pattern based optimization of transport capacity allocation |
US7333511B2 (en) * | 2002-08-29 | 2008-02-19 | Optimum Communications Services, Inc. | Dynamically channelizable packet transport network |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9699530B2 (en) | 2011-06-20 | 2017-07-04 | Plexxi Inc. | Optical architecture and channel plan employing multi-fiber configurations for data center network switching |
US9065582B2 (en) | 2011-06-20 | 2015-06-23 | Plexxi Inc. | Optical architecture and channel plan employing multi-fiber configurations for data center network switching |
US8842988B2 (en) | 2011-06-20 | 2014-09-23 | Plexxi Inc. | Optical junction nodes for use in data center networks |
US9876572B2 (en) | 2011-11-01 | 2018-01-23 | Plexxi Inc. | Configuring a computer network to satisfy multicast dispersion and latency requirements using affinity and network topologies |
US9301026B2 (en) | 2011-11-01 | 2016-03-29 | Plexxi Inc. | Affinity modeling in a data center network |
US9337931B2 (en) | 2011-11-01 | 2016-05-10 | Plexxi Inc. | Control and provisioning in a data center network with at least one central controller |
US9288555B2 (en) | 2011-11-01 | 2016-03-15 | Plexxi Inc. | Data center network architecture |
US9204207B2 (en) | 2011-11-01 | 2015-12-01 | Plexxi Inc. | Hierarchy of control in a data center network |
US9887777B2 (en) | 2011-11-01 | 2018-02-06 | Plexxi Inc. | Affinity modeling in a data center network |
US9942623B2 (en) | 2011-11-01 | 2018-04-10 | Plexxi Inc. | Data center network architecture |
US9450815B2 (en) | 2013-07-11 | 2016-09-20 | Plexxi Inc. | Network node connection configuration |
US9800472B2 (en) | 2013-07-11 | 2017-10-24 | Plexxi Inc. | Network node connection configuration |
US10063426B2 (en) | 2013-07-11 | 2018-08-28 | Plexxi Inc. | Network node connection configuration |
Also Published As
Publication number | Publication date |
---|---|
GB2473174A (en) | 2011-03-02 |
GB2473174B (en) | 2012-05-16 |
GB201100501D0 (en) | 2011-02-23 |
US20090310486A1 (en) | 2009-12-17 |
US8804760B2 (en) | 2014-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7986713B2 (en) | Data byte load based network byte-timeslot allocation | |
US8804760B2 (en) | Network data transport multiplexer bus with global and local optimization of capacity allocation | |
US7242686B1 (en) | System and method for communicating TDM traffic through a packet switch fabric | |
US8571062B1 (en) | Multichannel data conversion into packets | |
US6965619B2 (en) | Flexible multiplexer/demultiplexer and method for transport of optical line data to a wide/metro area link | |
US8848533B1 (en) | Time division multiplex packet fabric ingress scheduler and method | |
CN101151846B (en) | Method and system for shaping traffic in a parallel queuing hierarchy | |
EP2512078B1 (en) | Method and apparatus for mapping traffic using virtual concatenation | |
US11683739B2 (en) | Radio link aggregation | |
US20030048797A1 (en) | System and method for maximizing the traffic delivery capacity of packet transport networks via real-time trafffic pattern based optimization of transport capacity allocation | |
CN101146091B (en) | Multi-channel data output method and system | |
US20040042495A1 (en) | Dynamically channelizable packet transport network | |
CN101124756B (en) | System and method for multiplexing PDH and packet data | |
JP5418086B2 (en) | Transmission apparatus and signal transmission method | |
JP5320017B2 (en) | Transmission equipment | |
US7558260B2 (en) | Byte-timeslot-synchronous, dynamically switched multi-source-node data transport bus system | |
JP5610510B2 (en) | Wireless transmission apparatus and wireless transmission method | |
US6931023B2 (en) | Access device and method thereof for accessing a network | |
US6862295B1 (en) | Asynchronous scheduler | |
EP1936849B1 (en) | Method for mapping and demapping data information over the members of a concatenated group | |
JP2005244328A (en) | Wireless transmitter and transmitting method | |
US20210185562A1 (en) | Radio communication device, radio communication system, and radio communication method | |
Ducharme et al. | IP Frame Switching A Third-Generation SONET Technology | |
KR20110068736A (en) | Method and apparatus for switching packet/tdm including tdm circuit and carrier ethernet packet signal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09763138 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 6/DELNP/2011 Country of ref document: IN |
|
ENP | Entry into the national phase |
Ref document number: 1100501 Country of ref document: GB Kind code of ref document: A Free format text: PCT FILING DATE = 20090511 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1100501.4 Country of ref document: GB |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09763138 Country of ref document: EP Kind code of ref document: A1 |