37 results about "Terabit" patented technology

An optical transceiver system is provided that comprises multiple parallel transceiver modules that are mounted on a card. The transceiver card is small in terms of spatial dimensions, has very good heat dissipation characteristics, and is capable of simultaneously transmitting and receiving data at a rate equal to or greater than approximately one Tb per second (1 Tb / s). A plurality of the transceiver systems may be interconnected to achieve a communications hub system having even higher bandwidths. In addition, the transceiver system may be configured such that each card has a routing controller mounted thereon for performing router functions. The router functions include, for example: causing signals received by one transceiver module on the card to be routed to and transmitted by another of the transceiver modules; causing signals received by one transceiver module on the card to be retransmitted by the same transceiver module over one of it's optical transmit channels; and causing signals received by one transceiver module on the card to be routed to and transmitted by a transceiver module on a different card.

Embodiments of the present invention provide an optical network and switch architecture that provides non-blocking routing from an ingress router to an egress router in the network on a port-to-port basis. The present invention provides routing for fixed and variable length optical data packets of varying types (including Internet Protocol (IP), data, voice, TDM, ATM, voice over data, etc.) at speeds from sub-Terabit per second (Tbps), to significantly in excess of Petabit per second (Pbps). The present invention includes the functionality of both large IP routers and optical cross-connects combined with a unique, non-blocking optical switching and routing techniques to obtain benefits in speed and interconnected capacity in a data transport network. The present invention can utilize a TWDM wave slot transport scheme in conjunction with a just-in-time scheduling pattern and a unique optical switch configuration that provides for non-blocking transport of data from ingress to egress.One embodiment of the present invention includes a router comprising an ingress edge unit with one or more ports and an egress edge unit with one or more ports connected by a switch fabric. The ingress edge unit can receive optical data and convert the optical data into a plurality of micro lambdas, each micro lambda containing data destined for a particular egress edge port. The ingress edge unit can convert the incoming data to micro lambdas by generating a series of short-term parallel data bursts across multiple wavelengths. The ingress edge unit can also wavelength division multiplex and time domain multiplex each micro lambda for transmission to the switch fabric in a particular order. The switch fabric can receive the plurality of micro lambdas and route the plurality of micro lambdas to the plurality of egress edge units in a non-blocking manner. The router can also include a core controller that receives scheduling information from the plurality of ingress edge units and egress edge units. Based on the scheduling information, the core controller can develop a schedule pattern (i.e., a TWDM cycle) to coordinate the time domain multiplexing of micro lambdas at the plurality of ingress edge units and non-blocking switching of the micro lambdas at the switch fabric.

An extremely high resolution holographic lens is provided, comprising an interference pattern between light emitted from at least one tiny point source of light such as the tip of an optical fiber, having a diameter in the nanoscale region, and at least one plane or quasi-plane broad light beam. The holographic lens is employed for example in pickups for reading data off of, and writing data upon terabit optical storage mediums and nanoscale resolution microscopic media.

A perpendicular magnetic recording medium having sufficient perpendicular uniaxial magnetic anisotropy energy and a crystal grain size for realizing an areal recording density of one terabit or more per one square centimeter, and excellent in mass productivity, and a manufacturing method of the same are provided. On a substrate, a substrate-temperature control layer, an underlayer and a magnetic recording layer are sequentially formed. The magnetic recording layer is formed by repeating a magnetic layer stacking step N times (N>=2), which includes a first step of heating the substrate in a heat process chamber, and a second step of depositing, in a deposition process chamber, the magnetic recording layer constituted of an alloy mainly composed of FePt to which at least one kind of non-magnetic material selected from a group constituted of C and an Si oxide is added.

A method and apparatus are disclosed for scheduling arriving data packets for input to a switch having a plurality of input channels, and a plurality of output channels, the scheduling method is performed in successive scheduling phases where each scheduling phase further comprises at least log N scheduling iterations. The method is a parallelized weight-driven input queued switch scheduling algorithm which possesses good bandwidth and delay properties, is stable, and can be configured to offer various delay and quality of service (QoS) guarantees. The scheduling method utilizes envelope scheduling techniques and considers partially filled envelopes for scheduling.

A compact large density memory piezoactuated storage device and process for its fabrication provides an integrated microelectromechanical (MEMS) and / or nanoelectromechanical (NEMS) system and structure that features an integrated large density array of nanotips made of wear-resistant conductive ultrananocrystalline diamond (UNCD) in which the tips are actuated via a piezoelectric thin film integrated with the UNCD tips. The tips of the special piezoactuated storage device effectively contact an underlying metal layer (top electrode) deposited on a polarizable ferroelectric layer that is grown on top of another metal layer (bottom electrode) to form a ferroelectric capacitor. Information is imprinted in the ferroelectric layer by the polarization induced by the application of a voltage pulse between the top and bottom electrodes through the conductive UNCD tips. This integrated microelectromechanical (MEMS) and / or nanoelectromechanical (NEMS) system and structure can be efficiently used to imprint data in the ferroelectric layer for memory storage with high density in the gigabit (Gb) to terabit (Tb) range. An alternative memory media to the ferroelectric layer can be a phase change material that exhibits two orders of magnitude difference in electrical resistance between amorphous and crystalline phases.

A scheduling apparatus for a switch includes multiple schedulers which are assigned in a variety of ways to non-intersecting control domains for establishing connections through the switch. The control domains are defined by spatial and temporal aspects. The control domains may be dynamically selected and assigned to schedulers in a manner that achieves a high throughput gain. Control domains may be considered in a cyclic and / or a pipeline discipline for accommodating connection requests. The invention enables the realization of a highly scalable controller of a switching node of fine granularity that scales to capacities of the order of hundreds of terabits per second.

Most artificial neural networks are implemented electronically using graphical processing units to compute products of input signals and predetermined weights. The number of weights scales as the square of the number of neurons in the neural network, causing the power and bandwidth associated with retrieving and distributing the weights in an electronic architecture to scale poorly. Switching from an electronic architecture to an optical architecture for storing and distributing weights alleviates the communications bottleneck and reduces the power per transaction for much better scaling. The weights can be distributed at terabits per second at a power cost of picojoules per bit (versus gigabits per second and femtojoules per bit for electronic architectures). The bandwidth and power advantages are even better when distributing the same weights to many optical neural networks running simultaneously.

An optical transceiver system is provided that comprises multiple parallel transceiver modules that are mounted on a card. The transceiver card is small in terms of spatial dimensions, has very good heat dissipation characteristics, and is capable of simultaneously transmitting and receiving data at a rate equal to or greater than approximately one Tb per second (1 Tb / s). A plurality of the transceiver systems may be interconnected to achieve a communications hub system having even higher bandwidths. In addition, the transceiver system may be configured such that each card has a routing controller mounted thereon for performing router functions. The router functions include, for example: causing signals received by one transceiver module on the card to be routed to and transmitted by another of the transceiver modules; causing signals received by one transceiver module on the card to be retransmitted by the same transceiver module over one of it's optical transmit channels; and causing signals received by one transceiver module on the card to be routed to and transmitted by a transceiver module on a different card.

A switch module includes a switch integrated circuit (IC), a photonic integrated circuit (PIC), and a planar lightwave circuit (PLC). The PIC may include a plurality of light sources, an optical splitter, and a plurality of modulators. A dual MEMS may be used to align lens arrays, which may be used to couple light from the PIC to the PLC.

Wavelength division multiplexing (WDM) has enabled telecommunication service providers to fully exploit the transmission capacity of optical fibers. State of the art systems in long-haul networks now have aggregated capacities of terabits per second. Moreover, by providing multiple independent multi-gigabit channels, WDM technologies offer service providers with a straight forward way to build networks and expand networks to support multiple clients with different requirements. In order to reduce costs, enhance network flexibility, reduce spares, and provide re-configurability many service providers have migrated away from fixed wavelength transmitters, receivers, and transceivers, to wavelength tunable transmitters, receivers, and transceivers as well as wavelength dependent add-drop multiplexer, space switches etc. However, to meet the competing demands for improved performance, increased integration, reduced footprint, reduced power consumption, increased flexibility, re-configurability, and lower cost it is desirable to exploit / adopt monolithic optical circuit technologies, hybrid optoelectronic integration, and microelectromechanical systems (MEMS).

One embodiment of the present invention provides a switch. The switch includes a printed circuit board (PCB), a number of multi-channel optical transceivers mounted on the PCB, and a number of switch ports accessible from a front panel of the switch. The switch ports include a number of electrical interfaces that are electrically coupled to a switch chip mounted on the PCB, and a number of optical interfaces that are coupled to the switch chip via the multi-channel optical transceivers.

A switch module includes a switch integrated circuit (IC), a photonic integrated circuit (PIC), and a planar lightwave circuit (PLC). The PIC may include a plurality of light sources, an optical splitter, and a plurality of modulators. A dual MEMS may be used to align lens arrays, which may be used to couple light from the PIC to the PLC.