EP4008068A1 - Incrementally scalable, multi-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity - Google Patents
Incrementally scalable, multi-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivityInfo
- Publication number
- EP4008068A1 EP4008068A1 EP20848673.8A EP20848673A EP4008068A1 EP 4008068 A1 EP4008068 A1 EP 4008068A1 EP 20848673 A EP20848673 A EP 20848673A EP 4008068 A1 EP4008068 A1 EP 4008068A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- tier
- ntm
- ntms
- user
- ports
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/80—Optical aspects relating to the use of optical transmission for specific applications, not provided for in groups H04B10/03 - H04B10/70, e.g. optical power feeding or optical transmission through water
- H04B10/801—Optical aspects relating to the use of optical transmission for specific applications, not provided for in groups H04B10/03 - H04B10/70, e.g. optical power feeding or optical transmission through water using optical interconnects, e.g. light coupled isolators, circuit board interconnections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/15—Interconnection of switching modules
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/15—Interconnection of switching modules
- H04L49/1515—Non-blocking multistage, e.g. Clos
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/15—Interconnection of switching modules
- H04L49/1553—Interconnection of ATM switching modules, e.g. ATM switching fabrics
- H04L49/1569—Clos switching fabrics
-
- 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]
- H04L12/44—Star or tree networks
Definitions
- This invention relates to distributed, large scale communication systems comprised of fiber optic cables to transmit illumination and/or signals. More particularly, this invention relates to a multi-tiered robotically reconfigurable interconnection system comprised of large numbers of fiber optic cables aggregated into trunk lines connecting the multi -tiers and under software control.
- an incrementally scalable automated cross-connect system comprised of multiple modular, robotic interconnect units in a multi tier system.
- a two-tier design in which a first tier of robotic interconnect units has user ports and trunk ports in a predetermined ratio, a second tier of robotic interconnect units has trunk ports only, and trunk lines connecting the trunk ports of the first and second tiers of robotic interconnect units.
- the robotic interconnect units may individually consist of multi-interconnect modules enabling the number of interconnects in each unit to be increased, and the system enables the number of units within the system to be increased.
- An exemplary method of adding interconnections to an existing system of units in a non-service interrupting process is also disclosed.
- One general aspect includes a method of incrementally scaling a system of cross-connect units in a multi-tier arrangement to provide a given number of user
- the method also includes (a) for each particular network topology manager (NTM) of a plurality of NTMs in a first tier of said multi-tier arrangement: (a) connecting up to a particular set of k devices on said particular NTM in said first tier such that any device in said particular set can interconnect directly with any other device connected to said particular NTM, said particular NTM may include a plurality of interconnect modules, each module having a substantially identical number of interconnects.
- the method also includes (b) installing trunk line interconnections between said plurality of NTMs in said first tier to a number of NTMs in a second tier of said multi-tier arrangement. The method also includes where sufficient trunk line interconnections are installed to create inter-NTM
- Implementations may include one or more of the following features, alone or in various combination(s):
- each NTM in the second tier supports about 100 inter-NTM interconnections.
- each NTM in the second tier supports about 50 inter-NTM interconnections.
- Another general aspect includes a method of incrementally scaling a system of cross-connect units in a two-tier system of network topology managers (NTMs).
- the method also includes (a) connecting up to N/2 user ports to N/2 devices on a first NTM in said first tier, said first NTM having N user ports.
- the method also includes (b) adding an additional NTM to said first tier.
- the method also includes (c) installing additional fiber modules and/or an NTM in said second tier to support connections between NTM pairs in said first tier.
- Implementations may include one or more of the following features, alone or in various combination(s):
- the method where x % of local user connections may be different for each NTM in said first tier.
- N is about 1,000 to 2,000.
- N is 960 to 2,000
- M 4,800 to 160,000.
- Another general aspect includes a method of scaling a robotically
- the method also includes installing a first leaf NTM in first data center.
- the method also includes adding a second leaf NTM once x% of ports of first leaf NTM are connected to users in first data center, for some number x.
- the method also includes installing a spine NTM to connect (100-x)% of ports between first and second leaf NTMs in first data center and connecting spine NTM to leaf NTMs through trunk lines.
- the method also includes installing additional leaf NTMs in second data center and connecting this leaf NTM to the one or more spine NTMs in first data center.
- the method also includes repeating this process of adding leaf and spine NTMs and trunk lines therebetween as data centers are added.
- Implementations may include the method where x is 25 to 75.
- Another general aspect includes a method of incrementally deploying a fabric of passive interconnections.
- the method also includes deploying an interconnect fabric within a single rack and at least 100 user ports, where he capacity to increase the number of user ports is maintained by configuring no more than half the ports of each of said one or more first tier NTMs as user ports, and reserving the remaining ports of each of said one or more first tier NTMs as trunk ports.
- Implementations may include deploying (i) at least one additional NTM in said first tier and/or (ii) at least one additional NTM in a second tier.
- Another general aspect includes an incrementally scalable multi-tier NTM interconnect system.
- the incrementally scalable multi-tier NTM interconnect system also includes one or more tier 1 NTMs.
- the system also includes one or more tier 2 NTMs.
- the system also includes element managers for said NTMs to perform KBS routing of fiber.
- the system also includes trunk lines connecting tier 1 NTMs and tier 2 NTMs.
- the system also includes user interconnects connected to a portion of tier 1 NTM ports.
- the system also includes an NTM system controller accepting commands create an interconnection between a first user port and a second user port where said first user port and said second user port are on the one or more tier 1 NTMs, the controller in communication with all NTMs and sending reconfiguration instructions to all NTMs necessary to create an interconnection between said first user port and said second user port.
- Implementations may include one or more of the following features, alone or in various combination(s):
- an NTM trunk line routing mechanism determines an optimal set of NTMs based on a cost function to create an optimal fiber interconnection between said first user port and said second user port and passing through multiple NTMs and fiber trunk lines.
- the system where the cost function is designated to minimize one or more of: (i) insertion loss, and/or (ii) a number of hops through NTMs.
- Another aspect includes an NTM device in which a robot reconfigures an interconnect comprised of two optical fibers, each with a core and cladding, coextensive within a single element, to increase a number of user ports supported by a single tier 1 NTM device by a factor of two.
- Implementations may include one or more of the following features, alone or in various combination(s):
- the device where the single element has an outer diameter of about 0.4 to 0.5 mm.
- the device where the two optical fibers have cladding outer diameters of 50 to 80 microns.
- a method of incrementally scaling a system of cross-connect units in a multi tier arrangement to provide a given number of user interconnections comprising:
- trunk line interconnections between said plurality of NTMs in said first tier to a number of NTMs in a second tier of said multi-tier arrangement, wherein sufficient trunk line interconnections are installed to create inter- NTM interconnections required to support the given number of user interconnections and to enable any first user interconnection to connect to any second user interconnection.
- each NTM in the second tier supports about 100 inter-NTM interconnections.
- P5 The method of any of embodiment s) P1-P4, wherein the maximum capacity of user interconnections is equal to 5,000.
- P6 The method of any of embodiment s) P1-P5, wherein K is about 500.
- P7 The method of any of embodiment s) P1-P6 wherein at least some NTMs in the first tier are co-located.
- P8 The method of any of embodiment s) P1-P7, wherein at least some NTMs in the first tier are co-located with at least some NTMs in the second tier.
- P16 The method of any of embodiment s) P11-P15, wherein P is an integer multiple of P12.
- an interconnect fabric within a single rack and at least 100 user ports, wherein he capacity to increase the number of user ports is maintained by configuring no more than half the ports of each of said one or more first tier NTMs as user ports, and reserving the remaining ports of each of said one or more first tier NTMs as trunk ports.
- An incrementally scalable multi-tier NTM interconnect system comprising:
- tier 1 NTMs one or more tier 1 NTMs
- tier 2 NTMs one or more tier 2 NTMs
- an NTM system controller accepting commands create an interconnection between a first user port and a second user port , wherein said first user port and said second user port are on the one or more tier 1 NTMs, the controller in communication with at least some NTMs and sending reconfiguration instructions to at least some NTMs necessary to create an interconnection between said first user port and said second user port .
- an NTM trunk line routing mechanism determines an optimal set of NTMs based on a cost function to create an optimal fiber interconnection between said first user port and said second user port and passing through multiple NTMs and fiber trunk lines.
- An NTM device in which a robot reconfigures an interconnect comprised of two optical fibers, each with a core and cladding, coextensive within a single element, to increase a number of user ports supported by a single tier 1 NTM device by a factor of two.
- D34 The device of any of embodiment s) D32-D33, wherein the single element is terminated in a single connector with two adjacent cores.
- D35 The device of any of embodiment ⁇ s) D32-D34, wherein the two optical fibers have cladding outer diameters of 50 to 80 microns.
- FIGS. 1A-1K show aspects of exemplary scaling of Network Topology Managers (NTMs) according to exemplary embodiments hereof;
- FIGS. 2A-2W illustrate aspects of an exemplary process to vertically scale interconnects within a single NTM, and subsequently to scale horizontally interconnects across multiple NTMs;
- FIGS. 3A-3C depict aspects of an example interconnect system according to exemplary embodiments hereof.
- FIG. 4 illustrates an alternative system of NTMs according to exemplary embodiments hereof, scaling to 10,560 interconnects, with each NTM having 2,016 interconnects with any-to-any connectivity;
- FIG. 5 illustrates an alternative system of NTMs according to exemplary embodiments hereof, scaling to 9,600 interconnects, with each NTM-D having 2,016 interconnects with any A to any B connectivity;
- FIG. 6A illustrates aspects of an approach according to exemplary embodiments hereof to scaling system with duplex port NTM-D, in which the second tier is built out with partially populated simplex port NTM-S to provide any-to-any connectivity even though NTM-D are any A to any B;
- FIG. 6B illustrates aspects of the system of FIG. 6A scaled to 2,880 interconnections according to exemplary embodiments hereof;
- FIG. 6C illustrates aspects of the system of FIG. 6A scaled to 3,840 interconnections
- FIG. 6D illustrates aspects of the system of FIG. 6A scaled to 4,800 interconnections
- FIG. 6E illustrates aspects of the system of FIG. 6A scaled to 9,600 interconnections
- FIG. 7 is a block diagram of a system and controller according to exemplary embodiments hereof;
- FIG. 8 is a block diagram of aspects of an exemplary incremental deployment process of an NTM interconnect system according to embodiments hereof;
- FIGS. 9A-90 illustrate aspects of an example incremental build out according to exemplary embodiments hereof of an interconnect fabric serving an increasing number of data centers.
- FIG. 10 is an example of aspects of a 1+1 redundant, two-tier NTM
- KBS Knots, Braids and Strands
- NTM Network Topology Manager
- the term“mechanism” refers to any device(s), process(es), service(s), or combination thereof.
- a mechanism may be implemented in hardware, software, firmware, using a special-purpose device, or any combination thereof.
- a mechanism may be integrated into a single device or it may be distributed over multiple devices. The various components of a mechanism may be co-located or distributed. The mechanism may be formed from other mechanisms.
- the term“mechanism” may thus be considered to be shorthand for the term device(s) and/or process(es) and/or service(s).
- NTM Network Topology Manager
- NTM ports are classified herein as user ports (connected to external network devices) and trunk ports (connected to other NTMs to link NTMs from one tier to another, e.g., to link tier 1 and tier 2 NTMs).
- each NTM consists of multiple (e.g. 10) passive interconnect modules, each module substantially identical and containing a multiplicity (e.g. 48, 50, 96, 100, 120, or 192) of passive fiber interconnections.
- fiber optic interconnect devices are typically configured in multiples of 12 ports based on current industry standards; however, in some examples which follow, system examples in multiples of 10 ports will also be described for illustrative purposes.
- Passive fiber interconnect modules may be added to the NTM to create a larger non-blocking switch fabric, wherein an internal robot can move
- the number of ports may be increased from, for example, 48 interconnects to 5,280 interconnects and beyond by connecting multiple NTMs in a two- tier arrangement, wherein each NTM has a number of fiber interconnect modules necessary to support the required number of user ports and trunk ports.
- 10 NTMs each with about 1,008 interconnects in tier 1 may be joined into a fully non-blocking switch fabric through an additional 5 NTMs, each with about 960 trunk ports in tier 2.
- Tier 1 and Tier 2 NTMs may be connected with up to about 4,800 trunk line fibers.
- Each base NTM may include 48 x 48 interconnects, and up to ten 96 x 96 interconnect expansion modules may be added to the same unit.
- Example 1 Vertical Scalability from 48 to 1,008 Duplex Interconnects
- the NTM enables graceful scaling from 48 to 1,008 duplex fully non- blocking, any-to-any interconnections in single rack.
- Fiber modules may be added one on top of another within a common rack, and this is referred to here, for the purposes of description, as vertical scaling.
- Each fiber module has, for example, 96 interconnects (or alternatively, 48 or 192 interconnects).
- the NTM’s modular design with capacity for 1 to 10 fiber modules enables graceful scaling.
- FIGS. 1A-1K show aspects of scaling by stacking fiber modules vertically, one at a time, within a single mainframe, according to exemplary embodiments hereof.
- FIG. 1A depicts a base Network Topology Manager (NTM) system 100 with 48 LC duplex any-to-any interconnections 102.
- FIG. IB depicts the NTM system 100 with one 96 x 96 fiber module 104 installed, to increase to 144 LC duplex any-to-any
- FIG. 1C depicts the NTM system 100 with two 96 x 96 fiber modules 104, 106 installed, to increase to 240 LC duplex any-to-any interconnections.
- FIG. ID depicts the NTM system 100 with three 96 x 96 fiber modules 104, 106, 108 installed, to increase to 336 LC duplex any-to-any interconnections.
- FIG. IE depicts the NTM system 100 with four 96 x 96 fiber modules 104, 106, 108, 110 installed, to increase to 432 LC duplex any-to-any interconnections.
- FIG. IF depicts the NTM system 100 with five 96 x 96 fiber modules 112 installed, to increase to 528 LC duplex any-to-any interconnections.
- FIG. 1G depicts the NTM system 100 with six 96 x 96 fiber modules 114 installed, to increase to 624 LC duplex any-to-any interconnections.
- FIG. 1H depicts the NTM system 100 with seven 96 x 96 fiber modules 116 installed, to increase to 720 LC duplex any-to-any interconnections.
- FIG. II depicts the NTM system 100 with eight 96 x 96 fiber modules 118 installed, to increase to 816 LC duplex any-to-any interconnections.
- FIG. 1J depicts the NTM system 100 with nine 96 x 96 fiber modules 120 installed, to increase to 912 LC duplex any-to-any
- FIG. IK depicts the NTM system 100 with ten 96 x 96 fiber modules 122 installed, to increase to 1,008 LC duplex any-to-any interconnections.
- modules installed at a later time maintain full connectivity to any and all other modules within the mainframe. That is, there are no physical partitions limiting the ability to achieve any-to-any non-blocking connectivity.
- Any-to-any duplex cable (Tx, Rx fiber pair) connectivity may be achieved by connecting transmit lines to back (front) and receive lines to front (back), or vice versa.
- the transmit line of any user device may then be connected to the receive line of any other use device, or to the same user device in the case of a loopback.
- a single shared robot is able to reconfigure all connections within all modules of the NTM.
- An advantage of this vertical scaling design and process is that any interconnection may be established while consuming only two ports. Insertion loss is also minimized because any interconnection only passes through one pair of fiber optic connectors, the primary source of insertion loss in an NTM.
- NTMs may be deployed in a multi-tier interconnect architecture.
- two-tier designs are described, however, those of skill in the art will realize and understand, upon reading this description, that these concepts hold for three tiers, four tiers, etc. as well.
- the concepts described herein are applicable to N- tier interconnect architectures, for N > 2.
- 2A-2W illustrate aspects of an exemplary process to vertically scale from 48 to 528 interconnects within a single NTM, and subsequently to scale horizontally from 528 to 5,280 interconnects across multiple NTMs.
- Horizontal scaling refers herein to the process of connecting separate "leaf NTMs at the same tier through one or more "spine” NTMs at a different tier (e.g., connecting tier 1 or leaf NTMs (NTM Li) through tier 2 or“spine” NTMs (NTM Si)).
- FIG. 2A schematically illustrates a two-tier arrangement of NTMs according to exemplary embodiments hereof to achieve 5,280 non-blocking interconnects, with 10 NTMs in tier 1 and 5 NTMs in tier 2.
- FIG. 2B depicts an exemplary process of scaling up the number of
- interconnects according to embodiments hereof, beginning with a 48 interconnect minimally populated NTM 200-1.
- FIG. 2C depicts an exemplary process of scaling up the number of
- interconnects according to embodiments hereof, adding five modules to the NTM 200-1 to increase the number of interconnects to 528.
- FIG. 2D depicts an exemplary process of scaling up the number of
- interconnects according to embodiments hereof, adding a second NTM 200-2 with five modules in tier 1 to increase the number of interconnects to 1,056.
- FIG. 2E depicts an exemplary buildout of NTM tier 2 according to exemplary embodiments hereof, when it is necessary to interconnect the two-tier 1 NTMs 200-1, 200-2, starting with 96 interconnects between the two-tier 1 NTMs and adding one module each to the two-tier 1 NTMs, while not interrupting service on existing live interconnections and all NTMs nominally identical and based on the same platform.
- FIG. 2F depicts an exemplary process of scaling up in interconnects, adding a third NTM 200-3 with six modules in tier 1 to increase the number of interconnects to 1,584 and one module in tier 2, and adding 96 interconnections between it and tier 2.
- FIG. 2G depicts an exemplary process of scaling up in interconnects, adding a fourth NTM 200-4 with two modules in tier 1 to increase the number of interconnects to 1,680 and one module in tier 2, and adding 96 interconnections between it and tier 2.
- FIG. 2H depicts an exemplary process of scaling up in interconnects, adding four modules to the fourth NTM in tier 1 to increase the number of interconnects to 2, 112.
- FIG. 21 depicts an exemplary process of scaling up in interconnects, adding a fifth NTM 200-5 with four modules in tier 1 to increase the number of interconnects to 2,448.
- FIG. 2J depicts an exemplary process of scaling up in interconnects hereof, adding two modules to the fifth NTM in tier 1 to increase the number of interconnects to 2,640.
- FIG. 2K depicts an exemplary buildout of NTM tier 2 when it is necessary to interconnect tier 1 NTMs with additional interconnections, by adding a second tier 2 NTM 202-2 and cables, as necessary, in tier 2, to support more interconnections between NTMs in tier 1.
- FIG. 2L illustrates addition of fiber modules and cables to support more interconnections between NTMs in tier 1.
- FIG. 2M depicts addition of a third NTM 202-3 and cables in tier 2 to support more interconnections between NTMs in tier 1.
- FIG. 2N depicts addition of more modules and cables in tier 2 to support more interconnections between NTMs in tier 1.
- FIG. 20 depicts addition of a sixth NTM 200-6 in tier 1 to add 528 interconnections for a total of 3, 168, in addition to adding 192 interconnects between it and tier 2 NTM.
- FIG. 2P depicts addition of a seventh NTM 200-7 in tier 1 to add 528 interconnections for a total of 3,696 interconnections, in addition to adding 96 interconnects between it and tier 2 NTM.
- FIG. 2Q depicts an exemplary system of NTMs in which NTMs in tier 1 and
- NTMs tier 2 may be in different locations (e.g., different floors, buildings, cities, etc.) as shown by the dashed vertical in the drawing.
- tier 1 NTMs 200-1, 200-2, 200- 3, and 200-4 along with tier 2 NTMS 202-1, 202-2 are at a first location (e.g., the first floor of a building), while tier 1 NTMs 200-5, 200-6, and 200-7 along with tier 2 NTM 202-3 are at a second location distinct from the first location (e.g., the second floor of the building).
- NTMs may be co-located or in two or more distinct locations. NTMs in one tier may be co-located with NTMs in another tier and/or in distinct locations. (See also, e.g., FIG. 2U.)
- FIG. 2R illustrates addition of an eighth NTM 200-8 in tier 1 to add 528 interconnections for a total of 4,224 interconnections, and add 192 interconnects between it and tier 2 NTMs.
- FIG. 2S illustrates addition of a ninth NTM 200-9 in tier 1 to add 528 interconnections for a total of 4,752 interconnections, and add 288 interconnects between it and tier 2 NTMs.
- FIG. 2T illustrates addition of tenth NTM 200-10 in Tier 1 to add 528 interconnections for a total of 5,280 interconnections, and the addition of 288 interconnects between it and tier 2 NTMs.
- FIG. 2U illustrates deployment of individual NTMs of the system deployed on different floors, buildings, cities, etc.
- FIG. 2V illustrates addition of NTM 202-5 and cables in tier 2 to support more interconnects between tier 1 NTMs.
- FIG. 2W illustrates a full build out according to exemplary embodiments hereof of modules and cables in tier 2 to fully support interconnects between tier 1 NTMs, with 480 interconnects between each NTM of tier 1.
- tier 1 NTMs Li, L2, . . . L N may be configured so that all interconnections in tier 1 pass through some number of tier 2 NTMs Si, (i > 1), in practice a majority of interconnection paths remain within a single tier 1 device.
- the fraction of interconnections (defined herein as“local”) that remain within a tier 1 NTM is denoted by x%, and the fraction that pass to tier 2 (defined herein as“express”) is denoted by (100-x)% for some value x.
- Interconnects may be added in 96 interconnect increments to scale with network growth.
- any-to-any, non-blocking connectivity is important to eliminate physical partitioning of the interconnect fabric, because partitioning can add significant management complexity as it limits the ability of certain user ports to connect to other user ports in an automated fashion.
- partitioning can add significant management complexity as it limits the ability of certain user ports to connect to other user ports in an automated fashion.
- a 250-port cross-connect unit consisting of two separate, physically partitioned 125 port cross-connects would not allow fiber connections to be made between say, port 1 and port 250. Circumventing physical partitions would lead to the need for manual intervention and potentially service disrupting grooming.
- the system designs and scaling procedures presented herein overcome this problem.
- FIGS. 3A-3C depict aspects of an example interconnect system according to exemplary embodiments hereof.
- FIG. 3A depicts a 5,280-interconnect system according to exemplary embodiments hereof with 96 interconnects between each NTM of tier 1.
- FIG. 3B depicts a 5,280-interconnect system according to exemplary embodiments hereof with 288
- FIG. 3C depicts a 5,280-interconnect system according to exemplary embodiments hereof with a combination of 480, 384, 288, 192 and 96 interconnects between each NTM of tier 1.
- the multi-tier NTM system and method of scaling as disclosed herein provides full flexibility in terms of the number of inter-NTM connections (FIGS. 3A-3C), based on the number required for end users. Only the number of interconnections necessary to support projected user port growth over a certain time frame need to be installed initially. This flexibility is important because it enables upfront costs to be managed. It enables network operators to track their growth in fiber connectivity without requiring a massive overbuild of interconnections/cross-connections on day one. This provides a simple and compelling scaling approach without complex capacity planning and forecasting. In the particular examples illustrated in FIGS. 3A-3C, the scaling process is based on a few rules:
- interconnect modules in addition to 48 interconnects in the base system.
- the number of user ports may be increased from 528 to, for example, 720 user ports. This is advantageous because it reduces the tier 1 ports consumed by the trunk lines and it reduces the number of tier 2 ports connected to these trunk lines, thereby reducing the overall cost and footprint of the two- tier NTM interconnect fabric.
- This incremental scaling process provides for flexible buildout of tier 2 based on scaling requirements and inter-NTM demand to avoid blocking.
- tier 2 connections may be deployed as needed in blocks of 96 interconnects. This eliminates need to pre-deploy an excess number of interconnects.
- This particular example with 96 interconnect fiber modules scales gracefully to 5,280 user ports.
- the alternative examples above may be implemented to scale to 10,560, 21,020 and 42,040 user ports and beyond.
- Tables 1 and 2 below illustrate representative calculations for different example configurations. Note that parameter P should be rounded up to the nearest integer number of fibers.
- the individual native duplex NTM is any-A-to-any- B rather than the more general any-A-to-any-A, where A and B refer to a grouping of devices attached thereto.
- Any-A-to-any-A connectivity is the most general, without requiring that A devices can only connect to B devices.
- this requires a different scaling methodology compared to previous example, one that requires at least partial build-out of tier 2 on day 1 to provide the loopback interconnects that allow A devices to be connected to other A devices.
- FIG. 7 is a block diagram of a large, incrementally scalable NTM interconnect system 700 according to exemplary embodiments hereof.
- the system 700 comprises one or more tier 1 NTMs 702 and one or more tier 2 NTMs 704.
- Each NTM of each tier has an associated Element Manager 706-1, 706-2 (generally referred to as Element Manager 706) with KBS fiber routing engine 708-1, 708-2 (generally referred to as KBS Fiber Routing engine 708) to control the movement of a robot when moving a selected internal fiber interconnect.
- the NTM interconnect system 700 has an NTM System Controller 710 which selects and determines the connectivity of a trunk lines 712 within the NTMs at either end of each trunk line using the Trunk Line Routing Engine 714.
- the Trunk Line Routing Engine 714 determines an optimal path through the series of NTMs to minimize a user specified cost function (e.g. minimum insertion loss, minimum number of hops, minimum latency, minimum utilization, etc.).
- Reconfiguration instructions generated by the NTM System Controller 710 are send to the Element Managers 706 of the particular NTMs that must be reconfigured.
- a desired configuration of user ports 716 is input through the User Control Interface 718, which sends instructions to the NTM System Controller 710.
- FIG. 8 is a block diagram of an incremental deployment process of the interconnect fabric according to exemplary embodiments hereof, based on the parameters M, N, x, n, wherein M is the maximum user port capacity, N is the number of ports per NTM, x is the percentage of user ports that remain local to a single tier 1 NTM, and n is incremental number of user ports to be installed.
- a further example shows aspects of a design to increase the density of an individual NTM robotic cross-connect unit and thereby support additional user ports within a single unit, and by extension a system of such units.
- the capacity of an individual NTM unit is limited by its height and the vertical stacking height of optical fiber based on the outer diameter of the optical fiber internal to each fiber module.
- the nominal stacking height of each fiber at the internal one-dimensional backbone is 1 mm and for reduced form factor fiber this may be reduced to about 0.5 mm.
- the height of the internal one dimensional backbone of flexible fiber guide tubes (each tube with about 1 mm outer diameter) within the individual NTM unit is approximately 1,008 mm.
- the one-dimensional backbone may be made using smaller 0.5 mm outer diameter flexible guide tubes.
- the 1,008 mm backbone distance can then support up to 2,016 independent fibers.
- This unit is called the native simplex NTM with small form factor fiber and is illustrated in FIG. 4 .
- the width of the NTM is about 50 inches and the connector array at which the robot reconfigures is comprised of 24 rather than 12 columns.
- the small form factor optical fiber has, for example, an 80-micron outer diameter glass cladding and a 125-200-micron outer diameter polymer coated fiber.
- This system incorporates LC or other small form factor connectors.
- expansion fiber modules have twice the capacity for a given vertical height; for example, 192 x 192 interconnections within about 10 cm. Therefore, this NTM provides any-to-any, non-blocking connectivity with a factor of two increase in
- Example 4 Increased NTM Interconnect Density with Small Form Factor Duplex Fiber and Connectors
- the NTMs robot reconfigures native duplex fiber pairs instead of single fibers.
- Native duplex NTMs that is the NTM-D as, e.g., in FIG. 6A , refers to an NTM in which two fibers are placed within or extruded within a tube, so that any strand in the fiber module corresponds to two fibers instead of one fiber. This allows a duplex connection, which requires a Tx and Rx fiber pair, to replace the simplex or single fiber connection.
- the NTM-D can then be increased to twice the number of interconnects without increasing its physical size.
- a particular example uses reduced cladding optical fiber, wherein each fiber has an 80-micron cladding and 125-micron outer diameter polymer coating.
- the dual fibers are then extruded within a 400-micron tight buffer material, such as Hytrel, polyimide, or another suitable thermoplastic material with relatively low coefficient of friction and relatively high wear resistance.
- the output connector array to utilize small form factor duplex fiber connectors that fit within the same nominal size as LC simplex connectors.
- the two fibers of the duplex fiber pair may be terminated within one or two precision ferrules with polished end-faces of the small form factor connector.
- the“SN” small form factor connectors from Senko or the equivalent from US Conec achieve this size requirements.
- FIGS. 9A-90 illustrate aspects of a particular example of an incremental build out of the interconnect fabric serving an increasing number of data centers.
- This series of diagrams in FIGS. 9A-90 depict the two-tier NTM architecture, each diagram building on the prior diagram with the addition of new interconnects to illustrate the flexible evolution of the system as it grows.
- the interconnect fabric is non-blocking and any-to-any at all times. There is no need to make initial assumptions of the percentage of intra-building and percentage of inter-building cross-connections if each tier 1 NTM is filled with no more than 50% user ports, thereby reserving up to 50% of the ports for trunk lines.
- FIG. 9A is an example of a tier 1 NTM in an initial data center.
- FIG. 9B shows the tier 1 NTM of in an initial data center (FIG. 9A) with addition of user ports and tier 2 NTM.
- FIG. 9C shows the tier 1 NTM of FIG. 9B with the addition of user ports.
- FIG. 9D shows the example two-tier NTM architecture of FIG. 9C with addition of second data center to create a data center campus.
- FIG. 9E shows the example two-tier NTM architecture in a two data center campus (e.g., FIG. 9D) with addition of user ports.
- FIG. 9F shows a two-tier NTM architecture (e.g., FIG. 9E) with the addition of a third data center to the campus.
- FIG. 9G shows a two-tier NTM architecture in a three data center campus (e.g., FIG. 9F) with addition of user ports.
- FIG. 9H shows a two-tier NTM architecture (e.g., FIG. 9G) with the addition of a fourth data center to the campus.
- FIG. 91 shows a two-tier NTM architecture in a four data center campus (e.g., FIG. 9H) with addition of user ports.
- FIG. 9J shows a two-tier NTM architecture (e.g., FIG. 91) with the addition of a fifth data center to the campus.
- FIG. 9K shows a two-tier NTM architecture in a five data center campus (e.g., FIG. 9J) with addition of user ports.
- FIG. 9L shows a two-tier NTM architecture (e.g., FIG. 9K) with the addition of a sixth data center to the campus.
- FIG. 9M shows a two-tier NTM architecture in a six data center campus (e.g.,
- FIG. 9L with addition of user ports.
- FIG. 9N shows the two-tier NTM architecture (e.g., FIG. 9M) with the addition of a seventh data center to the campus.
- FIG. 90 shows the two-tier NTM architecture in a seventh data center campus (e.g., as shown in FIG. 9N) with addition of user ports.
- FIG. 10 is a further example of a robotic interconnect fabric for a data center campus, in which the two independent interconnect fabrics have substantially no
- FIG. 10 shows aspects of an exemplary 1+1 redundant, two-tier NTM architecture. This is typically referred to in the art as 1 : 1 redundancy and is a design that improves reliability and availability. If, for example, a large trunk cable between a pair of data centers is damaged, there is a separate trunk cable following a different physical path that would not likely remain operational. Discussion
- the NTM’ s modular construction of interconnect units, each with typically a hundred interconnects, enables graceful scaling vertically within a rack and scaling horizontally using an incrementally scalable, multi-tier interconnect fabric with user specified connectivity of user ports.
- the multi-NTM system is designed such that grooming or migration during the expansion process are eliminated. This ensures that there is no interruption of service while incrementally scaling; that is, all existing interconnects are unaffected by the installation of new interconnects across the NTMs, and are unaffected by the installation of additional trunk lines between the NTMs and dictated by capacity demands.
- the NTM system controller manages the complexity of the trunk lines and tier 2 NTMs so that users can specify the pair of user ports to be interconnected and the controller determines the robotic processes across the multiple NTMs optimal to interconnect the user specified ports.
- This automated interconnect fabric is non-blocking, allows any-to-any connectivity, and scales from 100 and 100K interconnects.
- Ethernet switches in leaf-spine, hub-spoke configuration, etc.
- these architectures do not directly apply to the unique nature of latching robotic physical interconnects.
- the fundamental difference is related to the orders of magnitude difference is reconfiguration time scale.
- Ethernet switches convert optical signals to electrical signals and route electronic data packets between ports on timescales of the order of 10 ps.
- the NTM moves physical fibers on the order of 2 minutes and during this time no signals can be transmitted.
- the physical fiber interconnects in the NTM system cannot necessary be groomed, nor can they be oversubscribed like an opto-electronic packet switch.
- process may operate without any user intervention.
- process includes some human intervention (e.g ., an act is performed by or with the assistance of a human).
- the phrase“at least some” means“one or more,” and includes the case of only one.
- the phrase“at least some ABCs” means“one or more ABCs”, and includes the case of only one ABC.
- term“at least one” should be understood as meaning“one or more”, and therefore includes both embodiments that include one or multiple components. Furthermore, dependent claims that refer to independent claims that describe features with“at least one” have the same meaning, both when the feature is referred to as“the” and“the at least one”.
- the term“portion” means some or all. So, for example,“A portion of Q” may include some of“Q” or all of“Q.”
- the phrase“using” means“using at least,” and is not exclusive.
- the phrase“using Q” means“using at least Q.”
- the phrase“using Q” does not mean“using only Q.”
- the phrase“based on” means“based in part on” or“based, at least in part, on,” and is not exclusive.
- the phrase“based on factor Q” means“based in part on factor Q” or“based, at least in part, on factor Q.”
- the phrase“distinct” means“at least partially distinct.” Unless specifically stated, distinct does not mean fully distinct. Thus, e.g, the phrase,“R is distinct from Q” means that“R is at least partially distinct from Q,” and does not mean that“R is fully distinct from Q.” Thus, as used herein, including in the claims, the phrase“R is distinct from Q” means that R differs from Q in at least some way.
- the terms“multiple” and“plurality” mean“two or more,” and include the case of“two”
- the phrase“multiple ABCs” means“two or more ABCs,” and includes“two ABCs.”
- the phrase“multiple PQRs” means“two or more PQRs,” and includes“two PQRs.”
- the present invention also covers the exact terms, features, values and ranges, etc. in case these terms, features, values and ranges etc. are used in conjunction with terms such as about, around, generally, substantially, essentially, at least etc. (i.e., "about 3” shall also cover exactly 3 or “substantially constant” shall also cover exactly constant).
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201962881908P | 2019-08-01 | 2019-08-01 | |
PCT/US2020/035776 WO2021021280A1 (en) | 2019-08-01 | 2020-06-02 | Incrementally scalable, multi-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity |
Publications (2)
Publication Number | Publication Date |
---|---|
EP4008068A1 true EP4008068A1 (en) | 2022-06-08 |
EP4008068A4 EP4008068A4 (en) | 2023-09-13 |
Family
ID=74228520
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP20848673.8A Pending EP4008068A4 (en) | 2019-08-01 | 2020-06-02 | Incrementally scalable, multi-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity |
Country Status (2)
Country | Link |
---|---|
EP (1) | EP4008068A4 (en) |
WO (1) | WO2021021280A1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018071341A2 (en) | 2016-10-10 | 2018-04-19 | Telescent Inc. | System of large scale robotic fiber cross-connects using multi-fiber trunk reservation |
US11252488B2 (en) | 2017-10-09 | 2022-02-15 | Telescent Inc. | Incrementally scalable, two-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity |
US11650598B2 (en) | 2017-12-30 | 2023-05-16 | Telescent Inc. | Automated physical network management system utilizing high resolution RFID, optical scans and mobile robotic actuator |
EP4025397A4 (en) | 2019-09-05 | 2023-10-25 | Telescent Inc. | Robotic fiber optic cross-connect systems, sub-systems, devices and methods |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
AU2003278712A1 (en) * | 2002-08-13 | 2004-02-25 | The Regents Of The University Of California | Compact wavelength-selective optical crossconnect |
WO2006054279A1 (en) * | 2004-11-17 | 2006-05-26 | Fiberzone Networks Ltd. | Optical crossbar switch |
US8463091B2 (en) * | 2007-10-15 | 2013-06-11 | Telescent Inc. | Methods to reconfigure all-fiber optical cross-connects |
US8203450B2 (en) * | 2008-01-02 | 2012-06-19 | Commscope, Inc. | Intelligent MPO-to-MPO patch panels having connectivity tracking capabilities and related methods |
US8621111B2 (en) * | 2010-09-22 | 2013-12-31 | Amazon Technologies, Inc. | Transpose box based network scaling |
WO2018071341A2 (en) * | 2016-10-10 | 2018-04-19 | Telescent Inc. | System of large scale robotic fiber cross-connects using multi-fiber trunk reservation |
-
2020
- 2020-06-02 EP EP20848673.8A patent/EP4008068A4/en active Pending
- 2020-06-02 WO PCT/US2020/035776 patent/WO2021021280A1/en unknown
Also Published As
Publication number | Publication date |
---|---|
WO2021021280A1 (en) | 2021-02-04 |
EP4008068A4 (en) | 2023-09-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11252488B2 (en) | Incrementally scalable, two-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity | |
WO2021021280A1 (en) | Incrementally scalable, multi-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity | |
US11579382B2 (en) | System of large scale robotic fiber cross-connects using multi-fiber trunk reservation | |
EP1981206B1 (en) | An exchange system and method for increasing exchange bandwidth | |
US8842988B2 (en) | Optical junction nodes for use in data center networks | |
Kabacinski | Nonblocking electronic and photonic switching fabrics | |
US8463091B2 (en) | Methods to reconfigure all-fiber optical cross-connects | |
AU2011305638B2 (en) | Transpose box based network scaling | |
CN101917331B (en) | Systems, methods, and apparatus for a data centre | |
JP6434533B2 (en) | Apparatus and method for scalable photonic packet architecture using PIC switches | |
US20140270762A1 (en) | System and method for data center optical connection | |
CN102571554A (en) | Methods and apparatus for forwarding-state transport in a distributed control plane | |
US11190860B2 (en) | Switch with a shuffle | |
US8284771B1 (en) | Run-time scalable switch fabric | |
US10587506B2 (en) | Video router | |
US20220132229A1 (en) | Incrementally scalable, two-tier system of robotic, fiber optic interconnect units enabling any-to-any connectivity | |
US6370295B2 (en) | Switch modules, a switch matrix including such modules, and a non-blocking modular switch network including such a matrix | |
US9059814B2 (en) | Modular optical patch panel | |
JP2019047160A (en) | Mpp network, construction method of mpp network, design device and design metho of mpp network | |
US10674625B1 (en) | Rack sideplane for interconnecting devices | |
CN112083530B (en) | NxM blocked low-cost optical cross matrix | |
EP4372439A1 (en) | Fabric modules for high-radix networks | |
US20240259717A1 (en) | Optical Interconnection Module Assembly for Spine-Leaf Network Scale-Out | |
US10116558B2 (en) | Packet switch using physical layer fiber pathways | |
Zhao et al. | Nested rings: a simple scalable ring-based ROADM structure for neural application computing in mega datacenters |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20220301 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
A4 | Supplementary search report drawn up and despatched |
Effective date: 20230811 |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: H04Q 11/00 20060101ALI20230807BHEP Ipc: H04B 10/80 20130101ALI20230807BHEP Ipc: G02B 6/28 20060101ALI20230807BHEP Ipc: H04B 10/27 20130101AFI20230807BHEP |