Optical communication system

Abstract

An optical communication system includes a controller configured to tune each of a plurality of transceivers to select a corresponding band of wavelengths from a wavelength spectrum, wherein the wavelength spectrum includes a plurality of subcarriers, wherein each of the plurality of subcarriers includes a corresponding and unique band of wavelengths obtained from the wavelength spectrum, wherein the plurality of transceivers are coupled to a first power splitter that receives the wavelength spectrum, wherein the power splitter is configured to generate a plurality of replicated wavelength spectra, wherein the each of the plurality of transceivers is configured to receive a corresponding one of the plurality of replicated wavelength spectra at a first power, wherein the first power splitter at a first power splitting stage receives the wavelength spectrum from a second power splitter at a second power splitting stage at a second power, wherein the second power is greater than the first power.

Description

[0001] This application is a divisional application of the application for invention patent application with the application date of March 30, 2022, application number 202280039585.0, and the invention name of “Optical Communication System”. TECHNICAL FIELD

[0002] The present disclosure relates to Internet Protocol (IP) switching fabric and / or topology, and more particularly to delivering the benefits of full IP Clos fabric and providing fast paths between the edge and leaf nodes of a data center using state-of-the-art flexible optical technology. BACKGROUND

[0003] In recent years, online services have been driving the allowance of online or cloud gaming between cloud gaming servers and clients connected through a network in a streaming format. The streaming format is gaining popularity due to the on-demand availability of game titles, the ability to execute more complex games, the ability to network between players for multiplayer gaming, asset sharing between players, instant experience sharing between players and / or spectators, allowing friends to watch friends play video games, letting friends join games that friends are playing, and the like.

[0004] Unfortunately, the demand is also pushing the limits of network connectivity capabilities. In particular, Internet Protocol (IP) fabrics that provide multi-stage circuit switched networks based on Clos switching topologies or fabrics have dominated the cloud industry in recent years due to their ability to scale by uniform elements and their highly flexible routing design. However, as 5-stage or even larger CLOS switching topologies become necessary to deliver the required host connectivity, the number of devices dedicated to “backbone” functions and without user / host connectivity has rapidly increased. These backbone dedicated devices add cost of switching hardware and transceivers, as well as additional operational complexity and active failure points that need to be managed.

[0005] It would be advantageous to reduce the capital, operational cost, and complexity of large switching topologies or fabrics, as well as also reduce latency between hosts and end user services. Embodiments of the present disclosure arise in this context. SUMMARY

[0006] Embodiments of the present disclosure relate to delivering the benefits of full Internet Protocol (IP) multi-stage switching fabric and also providing reduced latency without complexity and cost and also optical data center IP switching fabric (ODCF) and / or topology that provides fast paths between the edge and leaf nodes of a data center using state-of-the-art flexible optical technology.

[0007] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes a controller configured to tune each of a plurality of radial transceivers to select a corresponding band of wavelengths from a wavelength spectrum, wherein the wavelength spectrum includes a plurality of subcarriers, wherein each of the plurality of subcarriers includes a corresponding and unique band of wavelengths obtained from the wavelength spectrum, wherein the plurality of radial transceivers are coupled to a first power splitter that receives the wavelength spectrum, wherein the first power splitter is configured to generate a plurality of replicated wavelength spectra, wherein the each of the plurality of radial transceivers is configured to receive a corresponding one of the plurality of replicated wavelength spectra at a first power, wherein the first power splitter at a first power splitting stage receives the wavelength spectrum from a second power splitter at a second power splitting stage, wherein the second power is greater than the first power.

[0008] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes an axial optical transceiver configured to receive a wavelength spectrum at a total power, wherein the wavelength spectrum includes a plurality of subcarriers, wherein each of the plurality of subcarriers includes a corresponding and unique band of wavelengths obtained from the wavelength spectrum; a power splitter optically coupled to the axial optical transceiver and configured to receive the wavelength spectrum at the total power, wherein the power splitter is configured to generate one or more replicated wavelength spectra at a reduced power; a radial transceiver coupled to the power splitter and deployed within a rack assembly that serves a plurality of servers, wherein the radial transceiver is configured to receive a replicated wavelength spectrum at a power that is a portion of the total power; and a control system coupled to the radial transceiver, wherein the control system is configured to tune the radial transceiver to select a band of wavelengths from the replicated wavelength spectrum.

[0009] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes an axial optical transceiver configured to receive a wavelength spectrum at a total power, wherein the wavelength spectrum includes a plurality of subcarriers, wherein each of the plurality of subcarriers includes a corresponding and unique band of wavelengths obtained from the wavelength spectrum; a power splitting layer including a plurality of power splitters optically coupled to the axial optical transceiver and configured to output a plurality of replicated wavelength spectra; a plurality of radial transceivers coupled to the power splitting layer, each of the plurality of radial transceivers configured to receive a corresponding one of the plurality of replicated wavelength spectra at a corresponding power that is a portion of the total power; and a control system coupled to the plurality of radial transceivers, wherein the control system is configured to tune each of the plurality of radial transceivers to select a corresponding band of wavelengths from the corresponding one of the plurality of replicated wavelength spectra.

[0010] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes an axial optical transceiver configured to receive a wavelength spectrum. The optical communication system includes a power splitter coupled to the axial optical transceiver. The power splitter operates as a passive device configured to replicate the wavelength spectrum and output a plurality of replicated wavelength spectrums. Each of the plurality of replicated wavelength spectrums has a corresponding power that is a portion of a total power received from the axial optical transceiver. The optical communication system includes a plurality of radial transceivers coupled to the power splitter. Each of the plurality of radial transceivers is configured to receive a corresponding one of the plurality of replicated wavelength spectrums. Each radial transceiver is tunable to select a wavelength band that sets a bandwidth of the each radial transceiver. For example, a radial transceiver can be tuned to one or more wavelengths, where the plurality of tuned wavelengths can be contiguous or non-contiguous.

[0011] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes an axial optical transceiver configured to receive a wavelength spectrum. The optical communication system includes a power splitting layer coupled to the axial optical transceiver. The power splitting layer is configured to output a plurality of replicated wavelength spectrums. Each of the plurality of replicated wavelength spectrums has a corresponding power that is a portion of a total power received from the optical transceiver. The power splitting layer includes a plurality of power splitters configured in one or more cascaded layers, where each power splitter operates as a passive device. The optical communication system includes a plurality of radial transceivers coupled to the power splitting layer. Each of the plurality of radial transceivers is configured to receive a corresponding one of the plurality of replicated wavelength spectrums. Each radial transceiver is tunable to select a wavelength band that sets a bandwidth of the each radial transceiver. For example, a radial transceiver can be tuned to one or more wavelengths, where the plurality of tuned wavelengths can be contiguous or non-contiguous.

[0012] Embodiments of the present disclosure disclose an optical communication system. The optical communication system includes an axial optical transceiver configured to receive a wavelength spectrum. The optical communication system includes a first power splitter coupled to the axial optical transceiver. The first power splitter operates as a passive device configured to replicate the wavelength spectrum and output a first plurality of replicated wavelength spectrums. Each replicated wavelength spectrum of the first plurality of replicated wavelength spectrums has a corresponding power that is a portion of a total power received from the axial optical transceiver. The optical communication system includes a radial transceiver coupled to the first power splitter. The radial transceiver is configured to receive a replicated wavelength spectrum from the first plurality of replicated wavelength spectrums. The optical communication system includes a second power splitter coupled to the radial transceiver. The second power splitter is configured to receive the replicated wavelength spectrum from the first plurality of replicated wavelength spectrums and output a second plurality of replicated wavelength spectrums. The optical communication system includes a coherent transceiver of a server coupled to the second power splitter. The coherent transceiver is configured to receive a replicated wavelength spectrum from the second plurality of replicated wavelength spectrums. Additionally, the coherent transceiver is tunable to select a corresponding wavelength band from the replicated wavelength spectrum from the second plurality of replicated wavelength spectrums, where the wavelength band sets a bandwidth of the coherent transceiver. For example, the coherent transceiver can be tunable to one or more wavelengths, where multiple tuning wavelengths can be contiguous or non-contiguous.

[0013] Other aspects of the present disclosure will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, illustrated by way of example in the several figures of the drawing. BRIEF DESCRIPTION OF DRAWINGS

[0014] The present disclosure can best be understood by referring to the following description in conjunction with the accompanying drawings in which:

[0015] Figure 1 is a schematic diagram of a game cloud system for providing a game over a network between one or more compute nodes located at one or more data centers in accordance with an embodiment of the present disclosure.

[0016] Figure 2A is a schematic diagram of a plurality of rack assemblies including a plurality of compute nodes at a representative data center of a game cloud system in accordance with an embodiment of the present disclosure.

[0017] Figure 2B is a schematic diagram of a rack assembly including a top-of-rack switch configured to receive a full wavelength spectrum in accordance with an embodiment of the present disclosure.

[0018] Figure 3A shows an optical switching fabric with multipoint optics and configured with one super-spine in accordance with an embodiment of the present disclosure.

[0019] Figure 3B An optical exchange texture with multi-point optics and configured with one or more super-trunks is shown according to one embodiment of the present disclosure.

[0020] Figure 4A The optical exchange texture of a point-to-multipoint optics configuration according to one embodiment of the present disclosure is shown.

[0021] Figure 4B An embodiment according to this disclosure is shown. Figure 4A The connection between components of the optical data center exchange texture.

[0022] Figure 5A An exemplary layout of components of an optical switching texture for a data center according to one embodiment of the present disclosure is shown, the optical switching texture including a single splitter layer.

[0023] Figure 5B An optical switching texture of a data center according to one embodiment of the present disclosure is shown, the optical switching texture including multiple and cascaded splitter layers.

[0024] Figure 6A This is a schematic diagram illustrating a data center switching texture according to one embodiment of the present disclosure, the switching texture including the use of a 1x4 splitter across three cascaded splitter layers connected to a super backbone switching layer.

[0025] Figure 6B This illustrates an implementation of one embodiment according to the present disclosure. Figure 6A A schematic diagram of the physical data center layout with a switching texture.

[0026] Figure 7 This is a schematic diagram illustrating a data center switching texture according to one embodiment of the present disclosure, the switching texture being configured to perform switching at the host / server rack layer. Detailed Implementation

[0027] Although the following detailed description contains many specific details for illustrative purposes, those skilled in the art will understand that many variations and modifications to these details are within the scope of this disclosure. Therefore, aspects of this disclosure are set forth below without departing from the generality of the appended claims and without imposing any limitation on those claims.

[0028] Generally, embodiments of the present disclosure provide the advantages of delivering an all-Internet Protocol (IP) multi-level switching network and also provide an optical data center fabric (ODCF) and / or topology that uses fast paths delivered using flexible optical technology between the edge and leaf nodes of a data center. In particular, the middle switch layer can be replaced with passive optical power splitters in the data fabric. Advantages include reduced capital and operational costs for large optical data center fabrics, and increased simplicity of implementing optical data center fabrics. Additional advantages include ultra-low latency paths between hosts and end-user services, as latency of the middle layer is improved by using passive optical splitters instead of switches, as optical power splitters are purely passive optical devices that do not add latency substantially, whereas traditional switches perform optical-electrical-optical (OEO) conversions as well as some routing and / or switching operations (e.g., packet queuing, etc.), which introduce latency. That is, embodiments of the present disclosure significantly reduce latency through the optical data center fabric, while also reducing capital expenditures by removing one or more middle switch layers and replacing them with one or more middle power splitting layers (e.g., replacing expensive switch hardware with a less number of cost-effective passive optical devices in the fiber infrastructure), as well as reducing operational expenditures (e.g., reduced power usage and less maintenance required). Furthermore, embodiments of the present disclosure simplify the networking topology, resulting in easier scaling (e.g., expansion, multiple building implementations in one or more geographic locations, etc.) and fault diagnosis activities.

[0029] With the above general understanding of the various embodiments, exemplary details of the embodiments will now be described with reference to the various figures.

[0030] Throughout this specification, reference to an "application" or "game" or "video game" or "game application" or "game title" is intended to mean any type of interactive application that is directed by the execution of input commands. For illustrative purposes only, interactive applications include applications for gaming, word processing, video processing, video game processing, etc. Furthermore, the above-introduced terms are interchangeable.

[0031] Embodiments of the present disclosure are implemented within data centers that support many individual computing systems. Each of the computing systems can be configured to perform one or more functionalities in accordance with applications executing on the corresponding computing system. The computing systems within a particular data center can be configured to generally provide services (e.g., cloud gaming, cloud storage, etc.) primarily to clients, or can each be individually configured to provide unique services to clients. For example, a data center can be configured to generally provide cloud gaming services to many clients (e.g., remote computing devices each associated with a corresponding user). In another example, a data center can be configured to generally provide a financial services platform to many clients, such as those that facilitate high-frequency trading of financial securities (e.g., stocks, etc.). These data centers provided by embodiments of the present disclosure with features of low latency, reduced infrastructure cost, ease of scaling, and multi- building implementation are desirable for providing these various services. For example, for cloud gaming services, these data centers are well suited to provide the best user experience for players of complex games because there are few or no processing and / or display interruptions, and true on-demand gaming of any game title. Additionally, for financial services, these data centers are well suited to provide higher trading frequency and speed (i.e., fast response times when submitting trade orders, etc.) than more traditional configured data centers. For simplicity and ease of illustration, embodiments of the present disclosure are described in the context of data centers providing cloud gaming, but it should be understood that the data centers can be configured to provide other services and / or functionalities, such as financial services, etc.

[0032] Figure 1 is a schematic diagram of a system 100 for providing games between one or more computing nodes located at one or more data centers over a network 150 in accordance with an embodiment of the present disclosure, where the data centers are each configured using an optical data center IP switching fabric that uses passive optical power splitter devices, the optical data center IP switching fabric providing fast paths between edge and leaf nodes of the data center with reduced latency, simple implementation, and reduced cost. In accordance with an embodiment of the present disclosure, the system is configured to provide games between one or more cloud gaming servers over a network, and more specifically, for high speed access to network storage from pairs of computing nodes, such as within rack assemblies. Cloud gaming includes executing a video game at a server to generate video frames of a game render, which are then sent to a client for display. In other embodiments, the system 100 including one or more data centers can be configured to provide other services, such as financial services, etc.

[0033] It should also be understood that, in various implementations (e.g., in a cloud gaming environment or within a standalone system), cloud gaming can be performed using physical machines (e.g., central processing units (CPUs) and graphics processing units (GPUs)) or virtual machines, or a combination of both. For example, virtual machines (e.g., instances) can be created using a hypervisor (e.g., located in a data center) leveraging one or more components of the hardware layer (such as multiple CPUs, memory modules, GPUs, network interfaces, communication components, etc.). These physical resources can be arranged in racks (such as CPU racks, GPU racks, memory racks, etc.), where the physical resources in the rack can be accessed using top-of-rack switches that facilitate the fabrication of components for assembling and accessing instances (e.g., when building virtualization components for instances). Typically, the hypervisor can present multiple guest operating systems (OSS) with multiple instances configured with virtual resources. That is, each OS can be configured with a corresponding set of virtualization resources supported by one or more hardware resources (e.g., located in a corresponding data center). For example, each OS can be supported using a virtual CPU, multiple virtual GPUs, virtual memory, virtualized communication components, etc. Furthermore, instance configurations can be transferred from one data center to another to reduce latency. When saving a user's game session, user- or game-specific instant exploits can be utilized. Instant exploits can include any number of configurations described herein for optimizing the rapid rendering of video frames for the game session. In one implementation, game- or user-specific instant exploits can be transferred as configurable settings between data centers. The ability to transfer instant exploit settings enables efficient migration of games from data center to data center when users connect to play the game from different geographical locations.

[0034] System 100 includes a game cloud system 190 implemented via one or more data centers (e.g., data centers 1 to N). As shown, instances of the game cloud system 190 may reside in data center N, which provides management functionality, wherein the management functionality of the game cloud system 190 may be distributed through multiple instances of the game cloud system 190 in each data center. In some implementations, the management functionality of the game cloud system may reside outside of any data center.

[0035] This game cloud system 190 includes an allocator 191 configured to assign each of client devices (e.g., 1-N) to a corresponding resource in a corresponding data center. Specifically, when client device 110 logs into game cloud system 190, client device 110 can connect to an instance of game cloud system 109 at data center N, where data center N is geographically closest to client device 110. Allocator 191 is capable of performing diagnostic tests to determine the available transmit and receive bandwidth of client device 110. Based on the tests, allocator 191 can allocate resources to client device 110 very specifically. For example, allocator 191 can allocate a specific data center to client device 110. Furthermore, allocator 191 can allocate specific compute nodes to specific rack components, specific streaming arrays, or specific compute bays to client device 110. Allocation can be performed based on knowledge of the assets (e.g., games) available at the compute nodes. Previously, client devices were generally allocated to data centers without further allocation to rack components. In this way, allocator 191 can assign client devices requesting to execute a specific compute-intensive game application to compute nodes that are not currently running compute-intensive applications. Furthermore, load management of the allocation of compute-intensive game applications, such as those requested by clients, can be performed at allocator 191. For example, the same compute-intensive game application requested within a short period can be distributed across different compute nodes on different compute racks within a rack assembly or different rack assemblies to reduce the load on specific compute nodes, compute racks, and / or rack assemblies.

[0036] In some implementations, allocation can be performed based on machine learning. Specifically, resource demand can be predicted for a particular data center and its corresponding resources. For example, if it is predicted that a data center will soon handle many clients running compute-intensive gaming applications, allocator 191 can use this knowledge to allocate client device 110 and allocate resources that are currently not utilizing their full resource capacity. In another scenario, anticipating an increase in load at data center N, allocator 191 can switch client device 110 from the gaming cloud system 190 in data center N to available resources in data center 3. Furthermore, clients can be allocated resources in a distributed manner in the future, allowing resource load and demand to be distributed across the gaming cloud system, across multiple data centers, across multiple rack components, across multiple compute bays, and / or across multiple compute nodes. For example, client device 110 can be allocated resources from both the gaming cloud systems in data center N (e.g., via path 1) and data center 3 (e.g., via path 2).

[0037] Once client device 110 is assigned to a specific compute node in a corresponding compute rack of a streaming array, client device 110 connects to the corresponding data center via a network. That is, client device 110 can communicate with a data center (such as data center 3) that is different from the data center to which the assignment was performed.

[0038] According to one embodiment of this disclosure, system 100 provides games via a game cloud system 190, wherein the games are executed remotely from the client device (e.g., thin client) of the corresponding user playing the game. System 100 can provide game control to one or more users playing one or more games via a cloud gaming network or game cloud system 190 via network 150 in single-player or multi-player modes. In some embodiments, the cloud gaming network or game cloud system 190 may include multiple virtual machines (VMs) running on a host hypervisor, wherein one or more VMs are configured to utilize the hardware resources available to the host hypervisor to execute a game processor module. In some implementations, the game processor module may include a processing emulator to execute software applications configured for a processor or operating system to run on a processing emulator that may be configured with different processors and / or operating systems, wherein the processing emulator may or may not be virtualized. Network 150 may include one or more communication technologies. In some embodiments, network 150 may include fifth-generation (5G) network technology with advanced wireless communication systems.

[0039] In some implementations, wireless technologies can be used to facilitate communication. Such technologies may include, for example, 5G wireless communication technology. 5G is the fifth generation of cellular network technology. A 5G network is a digital cellular network in which the service area covered by the provider is divided into small geographical areas called cells. Analog signals representing voice and images are digitized in the phone, converted by an analog-to-digital converter, and transmitted as a bit stream. All 5G wireless devices in a cell communicate via radio waves with a local antenna array and low-power automatic transceivers (transmitters and receivers) in the cell through frequency channels allocated by the transceivers from a frequency pool reused from other cells. The local antennas are connected to the telephone network and the Internet via high-bandwidth fiber optic or wireless backhaul connections. As in other cellular networks, mobile devices moving from one cell to another automatically switch to the new cell. It should be understood that 5G networks are merely exemplary types of communication networks, and embodiments of this disclosure may utilize earlier generations of wireless or wired communication, as well as newer generations of wired or wireless technologies following 5G.

[0040] As shown in the figure, system 100, including game cloud system 190, can provide access to multiple game applications. Specifically, each client device may be requesting access to a different game application from a cloud gaming network. For example, game cloud system 190 can provide one or more game servers, which can be configured to run one or more virtual machines on one or more hosts to execute a corresponding game application. For example, the game server can manage virtual machines that support game processors that instantiate instances of game applications for users. Therefore, multiple game processors of one or more game servers associated with multiple virtual machines are configured to execute multiple instances of one or more game applications associated with games for multiple users. In this way, the backend server supports streaming media (e.g., video, audio, etc.) from multiple game applications to multiple corresponding users. That is, the game servers of game cloud system 190 are configured to stream data (e.g., rendered images and / or frames corresponding to the game) back to the corresponding client devices via network 150. In this way, computationally complex game applications can be executed at the backend server in response to controller input received and forwarded by the client devices. Each server is capable of rendering images and / or frames, then encoding (e.g., compressing) and streaming them to the corresponding client device for display.

[0041] In one implementation, the cloud gaming network or game cloud system 190 is a distributed game server system and / or architecture. Specifically, a distributed game engine executing game logic is configured as a corresponding instance of a game application. Generally, the distributed game engine employs each of the game engine's functionalities and distributes these functionalities for execution by numerous processing entities. Individual functionalities can be further distributed across one or more processing entities. Processing entities can be configured with different configurations, including physical hardware, and / or configured as virtual components or virtual machines, and / or configured as virtual containers, where a container differs from a virtual machine because it virtualizes an instance of a game application running on a virtualized operating system. Processing entities may utilize and / or rely on servers and their underlying hardware on one or more servers (compute nodes) within the cloud gaming network or game cloud system 190, wherein these servers may reside on one or more racks. The coordination, allocation, and management of the execution of these functionalities across various processing entities are performed by a distributed synchronization layer. Thus, the execution of these functionalities is controlled by the distributed synchronization layer to generate media (e.g., video frames, audio, etc.) for the game application in response to player controller input. The distributed synchronization layer enables these functions to be performed efficiently across distributed processing entities (e.g., through load balancing), allowing critical game engine components / functions to be distributed and reassembled for more efficient processing.

[0042] Figure 2AThis is a schematic diagram of multiple rack components 210 comprising multiple compute nodes at a representative data center 200A of a game cloud system according to one embodiment of this disclosure. For example, multiple data centers may be located worldwide, such as in North America, Europe, and Japan. In one embodiment, the rack components (e.g., configured for game streaming) are centered around compute nodes that run game applications, video games, and / or stream audio / video of game sessions to one or more clients.

[0043] Data center 200A includes multiple rack components 220 (e.g., rack components 220A to 220N). Each rack component includes a corresponding top-of-rack (TOR) switch and multiple compute bays. For example, a representative rack component 220N includes a top-of-rack switch 240N and multiple compute bays 230 (e.g., bays 230A to 230N). Other rack components may be configured similarly, with or without modification. In particular, each compute bay includes one or more compute nodes that provide hardware resources (e.g., processors, CPUs, GPUs, etc.). For example, compute bay 230N of the multiple compute bays 230 of rack component 220N is shown as including four compute nodes, but it should be understood that a rack component may include one or more compute nodes. Each rack component is coupled to a top-of-rack switch configured to provide communication with a management server 210 configured to manage the corresponding data center. The top-of-rack switch may be coupled to an optical data center texture configured to deliver data throughout the data center. For example, rack assembly 220N is coupled to top rack switch 240N. Top rack switch also provides communication to external communication networks (e.g., the Internet).

[0044] As shown in the figure, the management server 210 of data center 200A and ( Figure 1 The allocator 191 (shown) communicates to allocate resources to client devices 110. Specifically, the management server 210 can work with instance 190' of the game cloud system and with (e.g., Figure 1 The initial instance 190 of the game cloud system works in conjunction with the allocation of resources to client devices 110. In this implementation, the allocation is performed based on asset awareness, such as knowing what resources and bandwidth are needed and what resources and bandwidth are available at the data center. Therefore, for illustration, the implementation of this disclosure is configured to allocate client devices 110 to specific compute nodes 232 of corresponding compute racks 231 of corresponding rack components 220B.

[0045] Furthermore, the management server 210 of data center 200A includes a switch controller 215 that communicates with each of a plurality of top-of-rack (TOR) switches 240. As previously described, each rack component may be configured with a corresponding cluster or top-of-rack switch. For example, rack component 220A includes top-of-rack switch 240A, rack component 220B includes top-of-rack switch 240B, rack component 220C includes top-of-rack switch 240C, ..., and rack component 220N includes top-of-rack switch 240N. In particular, the switch controller 215 may configure each top-of-rack switch to be tuned to a selectable wavelength band, wherein the band sets the bandwidth of the transceiver of the corresponding rack component. In this way, each rack component may be reconfigured at various points in time depending on how the compute nodes in the corresponding rack component will be utilized. For example, during peak gaming hours, a rack component may be tuned to receive a wavelength band sufficient to support the game. During off-peak gaming hours (e.g., dark hours), the same rack component may be tuned to receive a different wavelength band. For example, during off-peak hours, rack components can be tuned to receive less bandwidth (i.e., compared to the bandwidth received during peak hours) to maintain the minimum connection required for gaming. In this way, additional bandwidth (i.e., bandwidth not used by the rack component) can be dynamically allocated to other rack components that are currently experiencing greater demand, such as rack components with compute nodes performing data backup or maintenance, running deep learning algorithms, etc.

[0046] Figure 2B This is a schematic diagram of a rack assembly according to one embodiment of the present disclosure, including a top-of-rack switch 250 configured to receive the full wavelength spectrum as part of an optical data center texture, wherein the top-of-rack switch 250 is further tuned to receive and communicate on discrete portions of the full wavelength spectrum (e.g., selected wavelength bands that set the bandwidth of the rack assembly). In this way, the optical data center texture uses one or more passive optical power splitter devices to provide a fast path between the edge of the data center (e.g., a super backbone switch) and the leaf nodes (e.g., a top-of-rack switch).

[0047] Specifically, rack assembly 220X can be configured to provide game streaming capabilities using compute nodes executing one or more game applications. In other implementations, compute nodes can be used to execute other types of applications. Rack assembly 220X may include a network storage device (not shown) configured to store game content (e.g., game logic, game applications, game data, user data, etc.) for high-speed access by one or more compute nodes and / or servers (280A to 280N). One or more compute nodes and / or servers may be configured as a streaming array. In other embodiments, the network storage device is located remotely from rack assembly 220X, such as within a distributed storage device configured to store multiple game applications (e.g., a complete supplement to a game portfolio for a game cloud system). As shown in the figure, Figure 2B The schematic diagram illustrates an advanced rack design for rack assembly 220X, which can represent Figure 2A One or more of a plurality of rack assemblies 220. For example, rack assembly 220X may represent rack assembly 220N.

[0048] Furthermore, rack assembly 220X includes a top-of-rack switch 250, which may also be referred to as a leaf node or leaf node switch. The top-of-rack switch 250 is configured to receive the full wavelength spectrum provided within an optical data center texture for data networking throughout the data center, and is further tuned to receive and communicate on discrete portions of the full wavelength spectrum. For example, the top-of-rack switch 250 receives the full wavelength spectrum from one or more passive optical power separation layers. Specifically, the top-of-rack switch 250 can be communicatively coupled to other compute nodes and / or servers within the same rack assembly or different rack assemblies via the optical data center texture. For example, the top-of-rack switch can be communicatively coupled to a communication network (e.g., the Internet) to provide network communication outside the rack assembly and / or data center.

[0049] As shown, the top-of-rack switch 250 includes a network interface card or controller (NIC) 260 configured to provide communication (e.g., via the full wavelength spectrum) between the rack assembly 220X and the optical data center texture. In one embodiment, the NIC 260 includes a coherent receiver 275X configured to tune the rack assembly 220X to select a wavelength band from the full wavelength spectrum received from the top-of-rack switch (e.g., selecting a wavelength band that sets the bandwidth of the top-of-rack switch 250). In one embodiment, the coherent receiver 275X is configured within a hot-swappable device 270. For example, the hot-swappable device may be a small form factor pluggable (SFP) hot-swappable network interface module 270 including a transceiver providing access to the optical data center texture, wherein the top-of-rack switch is configured as a transceiver. In another embodiment, the hot-swappable device 270 is a four-channel small form factor pluggable (QSFP) hot-swappable network interface module that includes a transceiver providing access to the optical data center texture, thereby providing higher speed data access and greater bandwidth, wherein the top-of-rack switch is configured as a radiating transceiver.

[0050] In another embodiment, each of the compute nodes and / or servers 280A-280N may optionally include a corresponding coherent receiver. For example, server 280A may include coherent receiver 275A, server 280B may include coherent receiver 275B, ..., and server 280N may include coherent receiver 275N. In this way, tuning to the wavelength band can be performed at each of the compute nodes and / or servers instead of at the top-of-rack switch 250 of the corresponding rack assembly 220X. Figure 7 A more detailed discussion of providing tuning at compute nodes and / or servers is provided in the document.

[0051] As previously described, the top-of-rack switch 250 can be controlled by a management server via a control path (not shown). For example, the top-of-rack switch 250 can be dynamically tuned to receive specific wavelength bands that set the bandwidth of the top-of-rack switch.

[0052] Figure 3AAn optical data center switching texture 300A according to one embodiment of the present disclosure is illustrated, having point-to-multipoint optics configured between a super-backbone optical transceiver (e.g., a switching device) 301A and a plurality of leaf nodes 320 (e.g., radial transceivers). The leaf nodes may be top-of-rack switches of a plurality of rack components in a data center. As shown, the super-backbone optical transceiver 301A is communicatively coupled to each of the plurality of leaf nodes 320 (e.g., leaf switches or top-of-rack switches), such that the optical data center switching texture 300A effectively provides a fast path for delivery between the edge of the data center (i.e., the axis optical transceiver 301A) and the leaf nodes using flexible optical technologies (e.g., passive optical power splitters).

[0053] Specifically, the optical data center switching texture 300A transmits data to leaf nodes using the full wavelength spectrum, rather than separating the full wavelength spectrum by frequency. This is achieved by using passive optics such as optical power splitters in one or more intermediate layers (not shown) instead of using switching equipment, as will be discussed below. Figures 4A-4B Further description. For example, the super-trunk optical transceiver 301A may be configured with multiple ports (e.g., more than two channels) for transmitting and receiving data. Depending on the number of leaf nodes, one or more intermediate layers may be inserted between the super-trunk optical transceiver 301A and the multiple leaf nodes 320. That is, as the number of leaf switches increases, a channel from the super-trunk optical transceiver 301A may be coupled to one or more passive optical power splitters for transmitting the full wavelength spectrum output by the super-trunk optical transceiver 301A to more than one leaf node.

[0054] In some implementations, an amplifier may be implemented between the super-trunk optical transceiver 301A and the multiple leaf nodes 320 to improve the signal-to-noise ratio of the transmitted signal, although given the typical data path lengths encountered within a data center, amplification is not expected to be necessary. That is, in a data center implementation, the existing power from the optics within the optical data center switching texture 300A should be high enough that an amplifier is not required, but an amplification stage may be implemented where needed or in some other use cases (such as providing amplification at the axis end).

[0055] As shown in the figure, a switch controller 215A can be configured within the super-backbone optical transceiver 301A. The switch controller 215A is communicatively coupled to each of the multiple leaf nodes 320 via control path 350. In this way, the switch controller 215A can tune each of the leaf nodes to receive and / or process wavelength bands of the full wavelength spectrum emitted from the super-backbone optical transceiver 301A within the optical data center switching texture 300A. In some cases, the switch controller 215A can be implemented within a management server 210, such as...Figure 2A As shown, the management server 210 can be configured within the super backbone optical transceiver 301A or located remotely from the backbone optical transceiver 301A, as previously described. For example, a coherent receiver at a corresponding leaf node can be tuned to receive and / or process wavelength bands that set the bandwidth of the leaf node.

[0056] The optical data center switching texture 300A includes a single superbackbone optical transceiver 301A that acts as the axis in a one-axis and multi-radial configuration, where the radials are leaf nodes. The superbackbone optical transceiver 301A can provide data across the entire wavelength spectrum. For illustration only, the entire spectrum can carry selectable wavelengths at specific data rates (e.g., 16, 32, or 64 gigabits per second, etc.) transmitting at 400 gigabits per second (400G). The entire wavelength spectrum can be further divided into subcarriers (e.g., 25G or any other number of subcarrier wavelengths) using various multiplexing techniques. In one embodiment, the entire wavelength spectrum is divided into 25G subcarrier wavelengths. In this way, each of the radial transceivers can be tuned to receive one or more 25G subcarrier wavelengths. For example, a radial transceiver can be tuned to receive 25G subcarrier wavelengths, or 50G subcarrier wavelengths, or 75G subcarrier wavelengths, or 100G subcarrier wavelengths, or any increment or multiple of 25G subcarrier wavelengths.

[0057] Figure 3B An optical data center switching texture 300B with multi-point optics and configured with one or more superbackbones is illustrated according to one embodiment of the present disclosure. Each superbackbones is communicatively coupled to multiple leaf nodes (e.g., radial transceivers or top-of-rack switches) 320. For example, each of the superbackbones axial optical transceivers (e.g., switching devices) 301A and 301B is communicatively coupled to multiple leaf nodes 320. The leaf nodes may be top-of-rack switches for multiple rack components in a data center. In this way, the optical data center switching texture 300B effectively provides a fast path for delivery between the edge of the data center (i.e., axial optical transceivers 301A and 301B) and the leaf nodes using flexible optical technologies (e.g., passive optical power splitters).

[0058] The coupling between each of the super backbone optical transceivers 301A and 301B in the optical data center switching texture 300B and the multiple leaf nodes 320 is similar to Figure 3A The super-trunk optical transceiver 301A is coupled to multiple leaf nodes 320. In this way, when using both super-trunk optical transceivers 301A and 301B to transmit data, Figure 3B The optical data center switching texture 300B can provide approximately twice the performance of Figure 3AThe optical data center switching fabric has an operating capacity of 300A (e.g., throughput, bandwidth, etc.). Purely for illustration, if... Figure 3A The optical data center switching texture 300A uses the super backbone optical transceiver 301A to provide a full wavelength spectrum of transmission at 400 gigabits per second (i.e., 400G). Figure 3B The optical data center switching texture 300B can be configured to use super backbone optical transceivers 301A and 301B to provide a full wavelength spectrum of transmission at 800 gigabits per second (i.e., 800G). Additionally, Figure 3B The optical data center switching texture 300B provides the same operating capacity as the optical data center switching texture 300A in Figure 3, but also provides backup or failover services. That is, when a data delivery path above one or more ports on the super backbone optical transceiver 301A fails, one or more ports on the super backbone optical transceiver 301B can be used to restore those data paths.

[0059] As shown in the figure, a switch controller 215B can be configured within the super-backbone optical transceiver 301B. The switch controller 215B is communicatively coupled to each of the multiple leaf nodes 320 via control path 355. In this way, the switch controller 215B can tune each of the leaf nodes to receive and / or process wavelength bands of the full wavelength spectrum emitted from the super-backbone optical transceiver 301B within the optical data center switching texture 300B. In some cases, the switch controller 215B can be implemented within a management server 210, such as... Figure 2A As shown, the management server 210 can be configured within the super-backbone optical transceiver 301B or located remotely from the axis optical transceiver 301B, as previously described. For example, a coherent receiver at a corresponding leaf node can be tuned to receive and / or process wavelength bands that set the bandwidth of the leaf node. The tuning of the leaf node is performed collaboratively between the switch controller 215A of the super-backbone local transceiver 301A and the switch controller 215B of the super-backbone local transceiver 301B.

[0060] Figure 4A An optical data center switching texture 400 with point-to-multipoint optics is shown according to one embodiment of the present disclosure. In particular, the optical data center switching texture uses point-to-multipoint optics to directly and communicatively couple axial optical transceivers 410 to a plurality of radial optical transceivers 430 (e.g., leaf switches) through one or more intermediate layers including one or more passive optical devices.

[0061] As shown in the figure, the optical data center switching texture 400 includes an axial optical transceiver 410 (e.g., a super backbone switch) configured to receive a wavelength spectrum, also referred to as the full wavelength spectrum processed through the data center. The full wavelength spectrum 420 delivers the total bandwidth for data delivery within the optical data center switching texture 400. For illustration only, the full spectrum can be transmitted at an optional data rate aggregated at 400 gigabits per second (e.g., 400 Gbps) or 800 gigabits per second (800 Gbps). The optical devices in the optical data center switching texture 400 can be further subdivided into subcarriers using various multiplexing techniques, such as multiple 25 Gbps wavelengths and / or channels, as previously described. For example, as... Figure 4A As shown, the full wavelength spectrum 420 can be divided into 32 different 25G wavelengths or wavelength bands.

[0062] Axial optical transceiver 410 is communicatively coupled to optical power splitter 460. That is, optical power splitters can be used to decompose the point-to-multipoint optics of an optical data center switching texture 400, providing latency reduction across the entire data center by eliminating one or more switching layers, including conventional switches. Since optical power splitters are purely passive optical devices that substantially do not add latency, they reduce latency, whereas conventional switches perform opto-optical (OEO) conversion and some routing and / or switching operations (e.g., packet queuing, etc.), thus introducing latency. Furthermore, optical power splitters are less complex and less costly than switches used at intermediate layers in conventional switching textures. For example, in some implementations, any standard optical power splitter can be used within the optical data center switching texture 400. Power splitter 460 operates as a passive device configured to replicate wavelength spectrum 420 and output multiple replicated wavelength spectra 420'. For example, multiple replicated wavelength spectra 420' include replicated wavelength spectra 420A, replicated wavelength spectra 420B, ..., and replicated wavelength spectra 420N.

[0063] In one implementation, each replicated wavelength spectrum comprises the full wavelength spectrum 420 (e.g., 800 GHz wavelength) emitted by the axial optical transceiver 410, but at a lower power. That is, each of the multiple replicated wavelength spectra 420' has a corresponding power as part of the total power received by the optical power splitter 460 from the axial optical transceiver 410. In one implementation, the power splitter divides the total power or the power received with the full wavelength spectrum or replicated full wavelength spectrum uniformly or non-uniformly. For example, the power splitter may divide the total power of the full wavelength spectrum uniformly or non-uniformly within the multiple replicated wavelength spectra.

[0064] The optical data center switching texture 400 includes multiple radiative transceivers 430 coupled to a power splitter. For example, each radiative transceiver is coupled to the power splitter 460 via a separate connection (e.g., fiber optic cabling). In one implementation, each radiative transceiver is a leaf switch or leaf node, such as a top-of-rack switch at each rack component of the corresponding data center. More specifically, each of the multiple radiative transceivers is configured to receive a corresponding one of multiple replicated wavelength spectra. That is, each radiative transceiver receives the full wavelength spectrum, such that the optical data center switching texture 400 transmits data to the radiative transceivers using the full wavelength spectrum, rather than separating the full wavelength spectrum by frequency. Therefore, the point-to-multipoint optics used between the axial optical transceiver 410 and each of the radiative optical transceivers separates or decomposes power, but still transmits the full wavelength spectrum. For example, a radiant optical transceiver 430A is configured to receive a copied wavelength spectrum 420A, a radiant optical transceiver 430B is configured to receive a copied wavelength spectrum 420B, ..., and a radiant optical transceiver 430N is configured to receive a copied wavelength spectrum 420N.

[0065] Multiple radiative optical transceivers 430 are tunable to set multiple bandwidths based on the total bandwidth provided by the full wavelength spectrum 420. More specifically, each radiative transceiver is tunable to select a wavelength band that sets the bandwidth of the corresponding radiative transceiver (i.e., selects a discrete portion of the corresponding full wavelength spectrum being replicated). For example, each radiative transceiver may include a leaf switch or top-of-rack switch deployed within a corresponding rack assembly serving multiple hosts and / or servers. This can be achieved by using a tunable coherent receiver in the radiative optical transceiver. In particular, each radiative transceiver includes an optical coherent receiver configured to divide the replicated wavelength spectrum (e.g., the full wavelength spectrum) into selectable wavelength bands, such that each radiative transceiver can be dynamically tuned to receive the selectable wavelength band that defines the bandwidth of the corresponding radiative transceiver. For example, each radiative transceiver includes a coherent receiver configured to select a wavelength band from a corresponding one of the multiple replicated wavelength spectra received by the radiative transceiver.

[0066] Specifically, each radiative transceiver can be tuned to at least one subcarrier wavelength (e.g., a 25 GHz wavelength), but can be tuned to multiple subcarriers to increase the bandwidth on a single connection, such as to the corresponding radiative transceiver (e.g., a top-of-rack switch). For example, radiative optical transceiver 430A includes a coherent receiver 440A tuned to its bandwidth to receive two 25 GHz subcarrier wavelengths for a total of 50 GHz, radiative optical transceiver 430B includes a coherent receiver 440B tuned to its bandwidth to receive one 25 GHz subcarrier wavelength for a total of 25 GHz, ..., and radiative optical transceiver 430N includes a coherent receiver 440N tuned to its bandwidth to receive four 25 GHz subcarrier wavelengths for a total of 100 GHz. In addition, another radiometric transceiver may include a coherent receiver tuned to receive any multiple of the 25 GHz subcarrier wavelength (e.g., 25 GHz, 50 GHz, 75 GHz, 100 GHz, 125 GHz, etc.) or tuned to receive any multiple of the subcarrier wavelength greater than or less than 25 GHz. That is, while some components may be tuned to select multiples of the 25 GHz increment, others may be tuned to select larger or smaller increments based on current technology or the generation of communication within the data center.

[0067] The optical data center switching texture 400 achieves enhanced performance by reducing latency between the axial optical transceiver 410 and each of the plurality of radial optical transceivers 430 by eliminating one or more intermediate switch layers. One or more switches in the one or more intermediate switch layers are replaced with point-to-multipoint optics as passive devices that do not require input power to operate. In one embodiment, the point-to-multipoint optics include passive optical power splitters. Depending on the number of radial optical transceivers required within the data center, one or more intermediate passive optical power splitter layers may be implemented, with each layer including one or more power splitters. In one embodiment, the multiple layers may include a cascaded layer of power splitters.

[0068] In one implementation, each of the axial optical transceiver 410, the optical power splitter 460, and each of the plurality of radial optical transceivers 430 may be hot-swappable. For example, as previously described, each device may be included within a QSFP form factor. In one implementation, one or more axial optical transceivers and one or more optical power splitters (i.e., configured as one or more cascaded layers) may be mounted within a rack assembly. As previously described, each of the radial optical transceivers may be mounted as a corresponding top-of-rack switch in a corresponding rack assembly.

[0069] In one implementation, a control system is coupled to multiple coherent receivers of multiple radiative transceivers, wherein the control system is configured to tune the coherent receivers of each radiative transceiver to select a corresponding wavelength band. For example, the control system may include a switch controller 215C configured within an axial optical transceiver 410. The switch controller 215C is communicatively coupled to each of the multiple radiative optical transceivers 430 via a control path 450. In this way, the switch controller 215C can tune each of the radiative optical transceivers to receive and / or process wavelength bands of the full wavelength spectrum emitted from the axial optical transceiver 410 within the optical data center switching texture 400. In some cases, the switch controller 215C may be implemented within a management server 210, such as... Figure 2A As shown, the management server 210 may be configured within the axial optical transceiver 410 or located remotely from the axial optical transceiver 410, as previously described. For example, a coherent receiver at a corresponding radiative optical transceiver may be tuned to receive and / or process wavelength bands that set the bandwidth of the radiative optical transceiver.

[0070] In one implementation, the optical data center switching texture 400 also allows for dynamic allocation of bandwidth to the radiating optical transceivers. As traffic demand changes over time, the optical data center switching texture 400 can be configured to increase and / or decrease capacity as needed. For example, an optical data center switching texture 400 used for a large business system may be heavily used during business hours, and therefore more bandwidth can be allocated to selected rack components and their corresponding hosts / servers during business hours or peak periods. During off-peak or dark periods, the bandwidth allocated to each rack component can be dynamically tuned to allow for dynamic allocation of bandwidth among rack components performing either heavier or lighter operations. For example, bandwidth can be tuned to allocate and / or reallocate to various servers / network segments across different rack components for data backup, management operations, machine learning algorithms, or other off-peak applications. Furthermore, software-defined networking (SDN) can be configured to automate bandwidth allocation, enabling dynamic and automatic allocation of bandwidth based on demand.

[0071] In some implementations, an amplifier may be implemented between the axial optical transceiver 410 and the plurality of radial optical transceivers 430 to improve the signal-to-noise ratio of the transmitted signal, although given the data length paths typically encountered within a data center, amplification is not expected to be necessary. That is, the amplifier may be coupled between the radial transceivers and the power splitter 460, wherein the amplifier is configured to amplify the power of the replicated wavelength spectrum received by the radial transceivers. Alternatively, an amplifier may be coupled between the axial optical transceiver 410 and the power splitter 460, wherein the amplifier is configured to amplify the power of the full wavelength spectrum received by the power splitter 460.

[0072] Figure 4BAn embodiment according to this disclosure is shown. Figure 4A The configuration described in the article is an optical data center switching texture 400 with point-to-multipoint optical devices. Figure 4A and Figure 4B The switching texture 400 shown is configured similarly. That is, the optical data center switching texture 400 uses point-to-multipoint optics to directly and communicatively couple axial optical transceivers 410 to multiple radial optical transceivers 430 (e.g., leaf switches) through one or more intermediate layers including one or more passive optical devices.

[0073] More specifically, Figure 4B The optical data center switching fabric 400 shown illustrates the connections between components. As illustrated, the connection between the axial optical transceiver 410 and the optical power splitter 460, and the connection between the power splitter 460 and each of the plurality of radial optical transceivers 430, can be achieved via fiber optic cabling, which can be simplex or duplex. In simplex fiber optic cabling, the signal flows in one direction (i.e., upstream or downstream). In duplex fiber optic cabling, the signal can flow in both directions (i.e., upstream and downstream).

[0074] In one embodiment, the connection between the axial optical transceiver 410 and the optical power splitter 460 is a simplex fiber optic cabling, requiring separate cabling for upstream and downstream data traffic. In another embodiment, the connection between the axial optical transceiver 410 and the optical power splitter 460 is a full-duplex fiber optic cabling for handling both upstream and downstream data traffic.

[0075] In one embodiment, the connection between the optical power splitter 460 and the corresponding radiating optical transceiver is a simplex fiber optic cabling, requiring separate cabling for upstream and downstream data traffic. In another embodiment, the connection between the optical power splitter 460 and the corresponding radiating optical transceiver is a full-duplex fiber optic cabling used to handle both upstream and downstream data traffic.

[0076] Figures 5A-5B This illustrates an example of optical splitting in a data center environment. Depending on the data center's purpose, capacity, and the variability of bandwidth provisioning for each rack component, various physical fiber optic infrastructure topologies can be used in different implementations. This is for illustrative purposes only. Figures 5A-5BSome examples are provided, but it should be understood that the topology is not limited to these examples. Regardless of the topology implemented, the optical data center switching texture of the embodiments of this disclosure allows for maximum flexibility with a minimal fiber count. This means that for the same number of rack components within the data center, the amount of fiber used is significantly reduced and the active switching infrastructure is significantly reduced (i.e., fewer passive optical devices are used compared to the number of active switches at one or more intermediate layers) when compared to conventional switching texture designs.

[0077] In particular, Figure 5A An optical switching texture 500A for a data center according to one embodiment of the present disclosure is shown, the optical switching texture including a single power splitter layer. The optical data center switching texture 500A uses point-to-multipoint optics to directly and communicatively couple a super backbone to multiple radial optical transceivers 550 located in multiple rack assemblies 530. As shown, the optical switching texture 500A includes axial optical transceivers, a power splitter layer, and multiple radial optical transceivers 550.

[0078] Specifically, the optical switching fabric 500A includes a superbackbone switching device (e.g., a switch, an axial optical transceiver, etc.) configured to receive a wavelength spectrum (e.g., a full wavelength spectrum) 420 of the total bandwidth for data delivery, wherein the full wavelength spectrum 420 has total power. The superbackbone switching device may be configured within a superbackbone rack assembly 501A, which also includes passive optics.

[0079] Furthermore, the optical switching texture 500A includes a power separation layer coupled to a superbackbone (e.g., an axial optical transceiver) within the rack assembly 501A. The power separation layer is configured to output multiple replicated wavelength spectra, each replicated wavelength spectrum having a corresponding power as a portion of the total power received from the superbackbone 501A. In one embodiment, the power separation layer includes multiple power splitters configured in one or more cascaded layers, each power splitter operating as a passive device and configured to replicate the received wavelength spectrum at reduced power. The power separation layer may also be included within the rack assembly 501A.

[0080] like Figure 5AAs shown, the power separation layer comprises a layer, and more specifically, a power splitter 460A configured as a 1x16 power splitter. That is, the full wavelength spectrum is divided into sixteen replicated wavelength spectra. Each replicated wavelength spectrum has a corresponding power as a portion of the total power received from the super backbone switching equipment, as previously described. In one embodiment, the portion associated with the replicated wavelength spectrum is uniformly distributed. In another embodiment, the portion associated with the replicated wavelength spectrum is non-uniformly distributed.

[0081] As shown in the figure, the power splitter 460 has multiple outputs (e.g., channels) 540A to 540P, such as 16 channels. Each output has a corresponding replicated wavelength spectrum. The outputs are coupled to multiple top-of-rack switches 550 in multiple rack assemblies 530. For example, channel 540A from the power splitter 460A is coupled to top-of-rack switch 550A of rack assembly 530A, channel 540B is coupled to top-of-rack switch 550B of rack assembly 530B, channel 540C is coupled to top-of-rack switch 550C of rack assembly 530C, channel 540D is coupled to top-of-rack switch 550D of rack assembly 530D, channel 540E is coupled to top-of-rack switch 550E of rack assembly 530E, channel 540F is coupled to top-of-rack switch 550F of rack assembly 530F, channel 540G is coupled to top-of-rack switch 550G of rack assembly 530G, and channel 540H... Channel 540I is coupled to the top-of-rack switch 550H of rack assembly 530H, channel 540J is coupled to the top-of-rack switch 550J of rack assembly 530J, channel 540K is coupled to the top-of-rack switch 550K of rack assembly 530K, channel 540L is coupled to the top-of-rack switch 550L of rack assembly 530L, channel 540M is coupled to the top-of-rack switch 550M of rack assembly 530M, channel 540O is coupled to the top-of-rack switch 550O of rack assembly 530O, and channel 540P is coupled to the top-of-rack switch 550P of rack assembly 530P. In this way, a super backbone switch port can connect to 16 leaf switches or top-of-rack switches. It should be understood that a switch port can connect to any number of top-of-rack switches depending on the number of power separation layers implemented.

[0082] Specifically, the optical data center switching texture 500A includes multiple radio transceivers (e.g., top-of-rack switches) coupled to a power separation layer. Each of the multiple radio transceivers is configured to receive a corresponding one of multiple replicated wavelength spectra, wherein each radio transceiver is tunable to select the wavelength band that sets the bandwidth of each radio transceiver. Figure 5AAs shown, multiple top-of-rack switches 530 are coupled to a power splitting layer or power splitter 460A. As previously described, each top-of-rack switch receives a replicated wavelength spectrum and can be further tuned to select a wavelength band that sets the bandwidth of the corresponding top-of-rack switch. This is achieved by using a corresponding coherent optical receiver at each top-of-rack switch, which is configured to divide the replicated wavelength spectrum into selectable wavelength bands that define the bandwidth of the corresponding top-of-rack switch.

[0083] As previously described, the control system is coupled to the coherent receivers of multiple top-of-rack switches 530, wherein the control system is configured to tune each top-of-rack switch to select a corresponding wavelength band. Although the control system is not in... Figure 5A As shown, but the control system may include a switch controller communicatively coupled to each coherent receiver to tune each top-of-rack switch to receive the corresponding wavelength band from the full wavelength spectrum 420.

[0084] Figure 5B An optical switching texture 500B for a data center according to one embodiment of the present disclosure is shown, the optical switching texture including multiple and cascaded splitter layers. The optical data center switching texture 500B uses point-to-multipoint optics to directly and communicatively couple a super backbone to multiple radial optical transceivers 560 located in multiple rack assemblies 530. As shown, the optical switching texture 500B includes axial optical transceivers, cascaded power splitter layers, and multiple radial optical transceivers 550.

[0085] Specifically, the optical switching texture 500B includes a superbackbone switching device (e.g., a switch, an axial optical transceiver, etc.) configured to receive a wavelength spectrum (e.g., a full wavelength spectrum) 420 of the total bandwidth for data delivery, wherein the full wavelength spectrum 420 has total power. The superbackbone switching device may be configured within a superbackbone rack assembly 501B, which also includes passive optical devices in a cascaded power separation layer.

[0086] Specifically, the optical switching texture 500A includes a cascaded power separation layer configured to output multiple replicated wavelength spectra, wherein each replicated wavelength spectrum has a corresponding power as a portion of the total power received from the superbackbone in the rack assembly 501B. In one embodiment, the cascaded power separation layer includes multiple power splitters configured in one or more cascaded layers, wherein each power splitter operates as a passive device and is configured to replicate the received wavelength spectrum at reduced power. The cascaded power separation layer may also be included within the rack assembly 501B.

[0087] As shown in the figure, the cascaded power splitting layer includes a first layer comprising a power splitter 460B configured as a 1x4 power splitter. That is, the full wavelength spectrum is split into four replicated wavelength spectra. For example, power splitter 460B provides a replicated wavelength spectrum as an output via channel 560, another replicated wavelength spectrum as an output via channel 561, another replicated wavelength spectrum as an output via channel 562, and yet another replicated wavelength spectrum as an output via channel 563. Each replicated wavelength spectrum provided as an output from power splitter 460B has a corresponding power as a portion of the total power received from the super backbone switching equipment, as previously described. In one embodiment, the portion associated with the replicated wavelength spectrum is uniformly distributed. In another embodiment, the portion associated with the replicated wavelength spectrum is non-uniformly distributed.

[0088] Specifically, the optical data center switching texture 500B includes multiple radio transceivers (e.g., top-of-rack switches) coupled to a cascaded power separation layer. Each of the multiple radio transceivers is configured to receive a corresponding one of multiple replicated wavelength spectra, wherein each radio transceiver is tunable to select the wavelength band that sets the bandwidth of each radio transceiver. Figure 5B As shown, multiple top-of-rack switches 530 are coupled to a cascaded power separation layer. As previously described, each top-of-rack switch receives a replicated wavelength spectrum and can be further tuned to select a wavelength band that sets the bandwidth of the corresponding top-of-rack switch. This is achieved by using a corresponding coherent optical receiver at each top-of-rack switch, which is configured to divide the replicated wavelength spectrum into selectable wavelength bands that define the bandwidth of the corresponding top-of-rack switch.

[0089] As shown in the figure, a cascaded power splitting layer is coupled to multiple top-of-rack switches 550 in multiple rack assemblies 530. Specifically, each output from power splitter 460B is coupled to a corresponding top-of-rack switch via a corresponding power splitter. For example, an output on channel 560 is coupled to power splitter 460C configured as a 1x4 power splitter, an output on channel 561 is coupled to power splitter 460D configured as a 1x4 power splitter, an output on channel 562 is coupled to power splitter 460E configured as a 1x4 power splitter, and an output on channel 563 is coupled to power splitter 460F configured as a 1x4 power splitter. Each output provides a replicated wavelength spectrum.

[0090] Furthermore, each of the power splitters 460C, 460D, 460E, and 460F is coupled to multiple top-of-rack switches of the corresponding rack assembly. As shown, power splitters 460C, 460D, 460E, and 460F can be installed in the corresponding rack assembly, but the power splitters can be located elsewhere, such as in another rack assembly. In particular, power splitter 460C provides four outputs via channels 560A, 560B, 560C, and 560D. Each channel is further communicatively coupled to the corresponding top-of-rack switch. For example, from power splitter 460C, channel 560A is coupled to top-of-rack switch 550A of rack assembly 530A, channel 560B is coupled to top-of-rack switch 550B of rack assembly 530B, channel 560C is coupled to top-of-rack switch 550C of rack assembly 530C, and channel 560D is coupled to top-of-rack switch 550D of rack assembly 530D. Additionally, from power splitter 460D, channel 561A is coupled to the top-of-rack switch 550E of rack assembly 530E, channel 561B is coupled to the top-of-rack switch 550F of rack assembly 530F, channel 561C is coupled to the top-of-rack switch 550G of rack assembly 530G, and channel 561D is coupled to the top-of-rack switch 550H of rack assembly 530H. Furthermore, from power splitter 460E, channel 562A is coupled to the top-of-rack switch 550I of rack assembly 530I, channel 562B is coupled to the top-of-rack switch 550J of rack assembly 530J, channel 562C is coupled to the top-of-rack switch 550K of rack assembly 530K, and channel 562D is coupled to the top-of-rack switch 550L of rack assembly 530L. Additionally, from power splitter 460F, channel 563A is coupled to the top-of-rack switch 550M of rack assembly 530M, channel 563B is coupled to the top-of-rack switch 550N of rack assembly 530N, channel 563C is coupled to the rack switch 550O of rack assembly 530O, and channel 563D is coupled to the top-of-rack switch 550P of rack assembly 530P.

[0091] As previously described, the control system is coupled to the coherent receivers of multiple top-of-rack switches 530, wherein the control system is configured to tune each top-of-rack switch to select a corresponding wavelength band. Although the control system is not in... Figure 5B As shown, but the control system may include a switch controller communicatively coupled to each coherent receiver to tune each top-of-rack switch to receive the corresponding wavelength band from the full wavelength spectrum 420.

[0092] Figure 6AThis is a schematic diagram illustrating a data center switching texture 600A according to one embodiment of the present disclosure, the switching texture including the use of a 1x4 splitter across three cascaded power splitter layers connected to a super backbone switching layer 610. Specifically, the optical data center switching texture 600A uses point-to-multipoint optics to directly and communicatively couple the super backbone switching layer (e.g., axial optical transceivers) to multiple radial optical transceivers (e.g., leaf switches, i.e., top-of-rack switches) via one or more intermediate layers including one or more passive optical devices.

[0093] As shown in the figure, the optical data center switching texture 600A includes a super backbone switching layer 610 configured to receive a wavelength spectrum, also referred to as the full wavelength spectrum processed through the data center. The full wavelength spectrum delivers the total bandwidth used for data delivery within the optical data center switching texture 600A. For illustration only, the full spectrum can be transmitted at selectable data rates of 400 gigabits per second (e.g., 400G) or 800 gigabits per second (800G). Optical devices in the optical data center switching texture 600 can be further subdivided into subcarriers using various multiplexing techniques, such as multiple 25G wavelengths, as previously described. For example, the full wavelength spectrum can be divided into 32 different 25G wavelengths or wavelength bands.

[0094] The optical switching texture 600A includes three cascaded power splitter layers coupled to a super backbone switching layer 610. For example, the cascaded power splitter layers and the super backbone switching layer may reside within a rack assembly in a data center. The three cascaded splitter layers include a first cascade stage, a second cascade stage, and a third cascade stage of power splitters. The three cascaded power splitter layers are configured to output multiple replicated wavelength spectra, each replicated wavelength spectrum having a corresponding power as a portion of the total power received from the super backbone switching layer 610. In one embodiment, the cascaded power splitter layer includes multiple power splitters configured in one or more cascaded layers, each power splitter operating as a passive device and configured to replicate the received wavelength spectrum at reduced power.

[0095] Specifically, the first cascade stage includes a single 1x4 power splitter 460G, which provides four outputs via channels. Each output provides a wavelength spectrum with a copy of the total power as part of the full wavelength spectrum provided by the super backbone switching layer 610.

[0096] The second cascade stage includes four power splitters, each being a 1x4 (1*4) power splitter, and each power splitter is coupled to power splitter 460G in the first cascade. For example, power splitter 460H receives a replicated wavelength spectrum from power splitter 460G in the first cascade stage via a channel and provides four additional outputs, each providing a replicated wavelength spectrum. Similarly, power splitter 460I receives a replicated wavelength spectrum from power splitter 460G in the first cascade stage via a channel and provides four additional outputs, each providing a replicated wavelength spectrum. Power splitter 460J receives a replicated wavelength spectrum from power splitter 460G in the first cascade stage via a channel and provides four additional outputs, each providing a replicated wavelength spectrum. Furthermore, power splitter 460K receives a replicated wavelength spectrum from power splitter 460G in the first cascade stage via a channel and provides four additional outputs, each providing a replicated wavelength spectrum.

[0097] The third cascade stage comprises sixteen power splitters 460L, 460M, 460N, 460O, ..., and 460X. Each power splitter is similarly configured, with each power splitter coupled to the output of one of the power splitters from the third cascade stage. As a representative example, power splitter 460H from the second cascade stage provides four outputs coupled to power splitters 460L, 460M, 460N, and 460O. For example, each of power splitters 460L through 460O receives a corresponding replicated wavelength spectrum from power splitter 460H via a corresponding channel.

[0098] Three cascaded power splitter layers are coupled to multiple radiative optical transceivers, such as a top-of-rack switch for multiple rack assemblies. Each of the multiple radiative transceivers is configured to receive a corresponding replicated wavelength spectrum, wherein each radiative transceiver receives the full wavelength spectrum, but at a reduced power compared to the total power output by the super-backbone switching layer 610 providing the full wavelength spectrum, as previously described. Therefore, each radiative transceiver is configured to divide the replicated wavelength spectrum (e.g., the full wavelength spectrum) into selectable wavelength bands using a corresponding optical coherent receiver. That is, each radiative transceiver can be dynamically tuned to receive selectable wavelength bands that define the bandwidth of the corresponding radiative transceiver.

[0099] Figure 6B This illustrates an implementation according to an embodiment of the present disclosure comprising multiple and cascaded power splitter layers. Figure 6AThis is a schematic diagram of the physical data center layout of the optical data center switching texture 600A. The optical data center switching texture 600B uses point-to-multipoint optics to directly and communicatively couple the super backbone switching layer to multiple radial optical transceivers 550 located in multiple rack assemblies 530. As shown, the optical switching texture 600B includes a super backbone switching layer 690 (e.g., backbone switches, axial optical transceivers, etc.), a cascaded power separation layer, and multiple radial optical transceivers 550.

[0100] The super backbone switching layer (e.g., switches, axial optical transceivers, etc.) is configured to receive the wavelength spectrum (e.g., full wavelength spectrum) of the total bandwidth used for data delivery, where the full wavelength spectrum has total power.

[0101] The optical switching texture 600B includes a cascaded power separation layer configured to output multiple replicated wavelength spectra, wherein each replicated wavelength spectrum has a corresponding power as a portion of the total power received from the superbackbone switching layer. In one embodiment, the cascaded power separation layer includes multiple power splitters configured in one or more cascaded layers, wherein each power splitter operates as a passive device and is configured to replicate the received wavelength spectrum at reduced power.

[0102] The first cascade stage 631 includes a power splitter 460P configured as a 1x4 power splitter. That is, the full wavelength spectrum is divided into four replicated wavelength spectra. For example, the power splitter 460P provides output through four channels, each carrying a replicated wavelength spectrum. Each replicated wavelength spectrum provided as the output from the power splitter 460P has a corresponding power as part of the total power received from the super-backbone switching layer, as previously described.

[0103] The second cascade stage 632 of the power splitter includes four power splitters, each configured as a 1x4 power splitter. For example, the second cascade stage 632 includes power splitters 460R, 460S, 460T, and 460U, each power splitter coupled to power splitter 460P in the first cascade stage 631. Each power splitter is coupled to the top-of-rack switch in a corresponding rack assembly.

[0104] For example, power splitter 460R receives the wavelength spectrum from power splitter 460P through a corresponding channel, and further provides an additional wavelength spectrum with lower power to each of the top-of-rack switches 550A, 550B, 550C, and 550D in the first rack assembly 551 through four corresponding channels. Additionally, power splitter 460S receives the wavelength spectrum from power splitter 460P through a corresponding channel, and further provides an additional wavelength spectrum with lower power to each of the top-of-rack switches 550E, 550F, 550G, and 550H in the second rack assembly 552 through four corresponding channels. Furthermore, power splitter 460T receives the wavelength spectrum from power splitter 460P through a corresponding channel, and further provides an additional wavelength spectrum with lower power to each of the top-of-rack switches 550I, 550J, 550K, and 550L in the third rack assembly 553 through four corresponding channels. In addition, the power splitter 460U receives the wavelength spectrum from the power splitter 460P through a corresponding channel, and further provides an additional wavelength spectrum with lower power to each of the rack top switches 550M, 550N, 550O and 550P in the fourth rack assembly 554 through four corresponding channels.

[0105] As previously described, each of the plurality of radiative transceivers 530 (e.g., top-of-rack switches) is configured to receive a corresponding one of a plurality of replicated wavelength spectra, wherein each radiative transceiver is tuned to select the wavelength band that sets the bandwidth of each radiative transceiver. Figure 6B As shown, multiple top-of-rack switches 530 are coupled to a cascaded power separation layer. As previously described, each top-of-rack switch receives a replicated wavelength spectrum and can be further tuned to select a wavelength band that sets the bandwidth of the corresponding top-of-rack switch. This is achieved by using a corresponding coherent optical receiver at each top-of-rack switch, which is configured to divide the replicated wavelength spectrum into selectable wavelength bands that define the bandwidth of the corresponding top-of-rack switch.

[0106] like Figure 6B As shown, optical data center switching textures can be implemented using repeatable steps. Specifically, the physical connectivity of each rack component remains identical, such as power connections (hookups) and fiber optic cables used to connect to the optical data center switching texture. Therefore, the pre-configured bandwidth can be dynamically tuned to suit each rack type as it is added to the optical data center texture. This simplifies data center architecture and streamlines the physical provisioning and cabling process, reducing the number of optical connections required.

[0107] In one implementation, a first cascade stage 631 and a second cascade stage 632 of one or more power splitters may be located within an overhead or underfloor cable tray 620. Fiber optic cables coupling the two cascade stages 631 and 632 to the top-of-rack switches in each of the rack assembly rows 551 to 554 may be routed through the cable tray 620. Specifically, a consistent fiber optic drop-in may be provided for each rack assembly in the rack assembly rows 551 to 554 during installation. As previously described, control of each rack assembly provides dynamic tuning of the bandwidth received at each corresponding rack assembly, allowing each rack assembly to be dynamically configured to receive selectable bandwidth at any given time.

[0108] Figure 7 This is a schematic diagram illustrating a data center switching texture according to one embodiment of the present disclosure, configured to perform switching at the host / server rack layer. The optical data center switching texture uses point-to-multipoint optics to directly and communicatively couple the super backbone switching layer to multiple radial optical transceivers located in multiple rack assemblies. Specifically, data delivery is shown at rack assembly 700, where instead of providing a coherent receiver at the top-of-rack switch of the corresponding rack assembly, the coherent receiver can be moved further down the distribution stack from the switching texture providing data to the rack assembly to the host / server generating traffic within one or more rack assemblies. As bandwidth demands increase and low-latency paths become more critical, a reduction in the number of steps between endpoints in the network can provide higher bandwidth to endpoints (e.g., hosts and / or servers).

[0109] For example, a top-of-rack switch 550' (e.g., a radial optical transceiver) may be optically configured with a coherent receiver, as previously described. However, the operations performed by the coherent receiver can be further pushed to each compute node of the rack assembly, as will be further described below.

[0110] Specifically, the top-of-rack switch 550' can be configured to receive replicated wavelength spectra from the power splitter. As previously described, the axial optical transceiver is configured to receive a wavelength spectrum (e.g., a full wavelength spectrum) that delivers the total bandwidth for data delivery, where the full wavelength spectrum has total power. The cascaded power splitting layer includes one or more power splitters, each configured to output a corresponding replicated wavelength spectrum, where each replicated wavelength spectrum has a corresponding power as a portion of the total power associated with the full wavelength spectrum provided by the axial optical transceiver. For example, the power splitter can be configured to replicate wavelength spectra and provide multiple replicated wavelength spectra.

[0111] As shown in the figure, top-rack switch 550' can provide output to another power splitter 460X via port 0, which can be configured as a 1x8 power splitter. Power splitter 460X provides a corresponding replicated wavelength spectrum as output to each of the eight servers and / or compute nodes (e.g., servers 0 to 7) in group 701 via corresponding channels 711A, 711B, ..., and 711H. Additionally, top-rack switch 550' can provide output to another power splitter 460Y via port 1, which can be configured as a 1x8 power splitter. Power splitter 460Y provides a corresponding replicated wavelength spectrum as output to each of the eight servers (e.g., servers 0 to 7) in group 702 via corresponding channels 712A, 712B, ..., and 712H. Additionally, top-rack switch 550' can provide output to another power splitter 460Z via port 2, which can be configured as a 1x8 power splitter. Power splitter 460Z provides a corresponding replicated wavelength spectrum as output to each of the eight servers (e.g., server 0 to server 7) in group 703 via corresponding channels 713A, 713B, ..., and 713H. Ports 3 to 7 of top-rack switch 550' can be similarly configured to provide the corresponding replicated wavelength spectrum output to another group of eight servers (not shown).

[0112] Furthermore, each server can be configured with a corresponding coherent transceiver. Therefore, the coherent transceiver of a server coupled to the power splitter can be configured to receive the replicated wavelength spectrum. Additionally, the coherent receiver can be tuned to select the corresponding wavelength band from the replicated wavelength spectrum, setting the coherent transceiver and correspondingly the bandwidth of the server. In other words, the server's coherent receiver is configured to divide the replicated wavelength spectrum into selectable wavelength bands that define the bandwidth of the corresponding server.

[0113] Therefore, this disclosure describes the advantages of delivering full Internet Protocol (IP) multi-level switching networks and also provides optical data center textures (ODCF) and / or topologies that deliver fast paths between the edge and leaf nodes of a data center using flexible optical technologies.

[0114] It should be understood that the various embodiments defined herein can be combined or assembled into specific implementations using the various features disclosed herein. Therefore, the examples provided are merely possible examples and are not limited to a variety of implementations that can be defined by combining various elements. In some examples, some implementations may include fewer elements without departing from the spirit of the disclosed or equivalent embodiments.

[0115] The embodiments of this disclosure can be practiced with various computer system configurations, including handheld devices, microprocessor systems, microprocessor-based or programmable consumer electronics, minicomputers, mainframe computers, and the like. The embodiments of this disclosure can also be practiced in distributed computing environments where tasks are performed via remote processing devices based on wired or wireless network links.

[0116] In light of the above embodiments, it should be understood that embodiments of this disclosure can employ various computer-implemented operations involving data stored in a computer system. These operations are those that require physical manipulation of physical quantities. Any operation described herein that forms part of embodiments of this disclosure is a useful machine operation. Embodiments of this disclosure also relate to devices or apparatuses for performing these operations. Such devices may be specifically constructed for the desired purpose, or the devices may be general-purpose computers selectively activated or configured by computer programs stored in a computer. In particular, various general-purpose machines may be used with computer programs written in accordance with the teachings of this document, or it may be more convenient to construct more specialized devices to perform the desired operations.

[0117] This disclosure can also be implemented as computer-readable code on a computer-readable medium. A computer-readable medium is any data storage device that can store data that can subsequently be read by a computer system. Examples of computer-readable media include hard disk drives, network attached storage devices (NAS), read-only memory, random access memory, CD-ROMs, CD-Rs, CD-RWs, magnetic tapes, and other optical and non-optical data storage devices. Computer-readable media may include computer-readable tangible media distributed across network-coupled computer systems, enabling the distributed storage and execution of computer-readable code.

[0118] Although the method operations are described in a specific order, it should be understood that other housekeeping operations may be performed between operations, or operations may be adjusted so that they occur at slightly different times, or they may be distributed in a system that allows processing operations to occur at various intervals associated with the processing, as long as the processing of the covered operations is performed in the desired manner.

[0119] While the foregoing disclosure has been described in considerable detail for the purposes of clarity, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. Therefore, embodiments of the invention are to be considered illustrative rather than restrictive, and embodiments of this disclosure are not limited to the details given herein, but may be modified within the scope of the appended claims and their equivalents.

Claims

1. An optical communication system, comprising: The controller is configured to dynamically tune each of a plurality of radiative transceivers to select a corresponding wavelength band defining the bandwidth of the radiative transceiver from a wavelength spectrum, wherein the wavelength spectrum comprises a plurality of subcarriers, and each of the plurality of subcarriers comprises a corresponding and unique wavelength band obtained from the wavelength spectrum. The plurality of radiative transceivers are coupled to a first power splitter that receives the wavelength spectrum, wherein the first power splitter is configured to generate a plurality of replicated wavelength spectra, and each of the plurality of radiative transceivers is configured to receive a corresponding one of the plurality of replicated wavelength spectra at a first power. The first power separator in the first power separation stage receives the wavelength spectrum from the second power separator in the second power separation stage at a second power, wherein the second power is greater than the first power.

2. The optical communication system of claim 1, wherein the controller is configured to tune each of the plurality of radiative transceivers to select a corresponding wavelength band from one of the plurality of replicated wavelength spectra.

3. The optical communication system according to claim 2, wherein each of the plurality of radiant transceivers comprises: A coherent receiver coupled to the controller. The coherent receiver can be tuned by the controller to select the corresponding wavelength band from one of the plurality of replicated wavelength spectra.

4. The optical communication system of claim 1, wherein the replicated wavelength spectrum received by the radiative transceiver has a power that is a part of the total power of the wavelength spectrum received from the axial optical transceiver, the axial optical transceiver being configured to provide the wavelength spectrum to power separators in the first power separation stage and the second power separation stage.

5. The optical communication system of claim 1, wherein the first power splitter uniformly or non-uniformly divides the second power of the wavelength spectrum received from the second power splitter within the plurality of replicated wavelength spectra.

6. An optical communication system, comprising: An axial optical transceiver is configured to receive a wavelength spectrum at total power, wherein the wavelength spectrum comprises a plurality of subcarriers, each of the plurality of subcarriers comprising a corresponding and unique wavelength band obtained from the wavelength spectrum; A power splitter, optically coupled to the axial optical transceiver and configured to receive the wavelength spectrum at the total power, wherein the power splitter is configured to generate one or more replicated wavelength spectra at reduced power; A radiative transceiver, coupled to the power splitter and deployed within a rack assembly serving multiple servers, wherein the radiative transceiver is configured to receive a replicated wavelength spectrum as part of the total power. as well as A control system coupled to the transceiver, wherein the control system is configured to dynamically tune the transceiver to select a wavelength band defining the bandwidth of the transceiver from the replicated wavelength spectrum.

7. The optical communication system according to claim 6, further comprising: The coherent receiver in the network interface of the transceiver. The coherent receiver is coupled to the control system, and The coherent receiver can be tuned by the control system to select the wavelength band from the replicated wavelength spectrum.

8. The optical communication system of claim 7, wherein the coherent receiver is a hot-swappable device.

9. The optical communication system of claim 6, wherein the replicated wavelength spectrum received by the radiative transceiver has a power that is a portion of the total power of the wavelength spectrum received from the axial optical transceiver.

10. The optical communication system according to claim 9, further comprising: An amplifier, coupled between the transceiver and the power splitter, is configured to generate the power of the replicated wavelength spectrum.

11. The optical communication system of claim 6, wherein one or more of the axial optical transceiver, the power splitter, and the radiative transceiver are four-channel small form factor pluggable (QSFP) devices.

12. The optical communication system according to claim 6, further comprising: Multiple radiant transceivers are coupled to the power splitter and the control system. These transceivers are deployed within multiple rack assemblies, each rack assembly serving a corresponding set of servers. Each of the plurality of transceivers is configured to receive a corresponding replicated wavelength spectrum from the one or more replicated wavelength spectra generated by the power splitter, and The control system is configured to tune each of the plurality of transceivers to select a corresponding wavelength band from the corresponding replicated wavelength spectrum.

13. An optical communication system, comprising: An axial optical transceiver is configured to receive a wavelength spectrum at total power, wherein the wavelength spectrum comprises a plurality of subcarriers, each of the plurality of subcarriers comprising a corresponding and unique wavelength band obtained from the wavelength spectrum; A power separation layer includes multiple power separators optically coupled to the axial optical transceiver and configured to output multiple replicated wavelength spectra; Multiple radiative transceivers are coupled to the power separation layer, each of the multiple radiative transceivers being configured to receive a corresponding one of the multiple replicated wavelength spectra at a corresponding power that is part of the total power; as well as A control system coupled to the plurality of radiative transceivers, wherein the control system is configured to dynamically tune each of the plurality of radiative transceivers to select a corresponding wavelength band defining the bandwidth of the radiative transceiver from one of the plurality of replicated wavelength spectra.

14. The optical communication system of claim 13, wherein each of the plurality of power splitters is configured to replicate the wavelength spectrum of the power received at reduced power.

15. The optical communication system of claim 13, wherein the power separation layer is configured in one or more cascaded layers, the one or more cascaded layers comprising: A first power splitter is coupled to the axial optical transceiver and configured to replicate the wavelength spectrum and output a plurality of first replicated wavelength spectra; as well as A second power splitter is coupled to the first power splitter and configured to replicate the plurality of first replicated wavelength spectra and output a plurality of second replicated wavelength spectra.

16. The optical communication system of claim 13, wherein each of the plurality of radiant transceivers comprises: A coherent receiver coupled to the network interface of the control system. The coherent receiver can be tuned by the control system to select the corresponding wavelength band from one of the plurality of replicated wavelength spectra.

17. The optical communication system of claim 16, wherein the coherent receiver is a hot-pluggable device.

18. The optical communication system according to claim 13, Each of the plurality of replicated wavelength spectra has a corresponding power that is a portion of the total power of the wavelength spectrum received from the axial optical transceiver.

19. The optical communication system according to claim 13, further comprising: An amplifier, coupled between the radiative transceiver and the power splitter of the power separation layer, is configured to generate power of a replicated wavelength spectrum received by the radiative transceiver.

20. The optical communication system of claim 13, wherein one or more of the axial optical transceiver, the plurality of power splitters, and the plurality of radial transceivers are four-channel small form factor pluggable (QSFP) devices.

Images