Mellanox Technologies Tlv Ltd.

Mellanox Technologies, Ltd.

Barak Gafni

Aviv Kfir

Benny Koren

Gil Levy

Liron Mula

George Elias

Niv Aibester

Zachy Haramaty

Eitan Zahavi

Itamar Rabenstein

Henning Lysdal

Matty Kadosh

Shachar Raindel

Ido Bukspan

Sagi Kuks

Ilya Vershkov

Dror Goldenberg

Amir Roitshtein

Eyal Srebro

Roy Kriss

David Mozes

Shachar Bar Tikva

Avner Hadash

Ortal Ben Moshe

Eran Rubinstein

Elad Mentovich

Noam Bloch

Liran Liss

Paraskevas Bakopoulos

Idan Burstein

Lion Levi

Ioannis (Giannis) Patronas

Ariel Almog

Ran Ravid

Lavi Koch

Amiad Marelli

Freddy Gabbay

Alex Shpiner

Tamir Ronen

Oded Zemer

Noam Katz Abramovich

Amit Kazimirsky

Zvi Rechtman

Maty Golovaty

Eran Gil Beracha

Jacob Ruthstein

Gal Shohet

Harold Rosenstock

Ortal Ben-Moshe

Linor Nehab

Zachy Hamamaty

Shmuel Shichrur

Adam V Richards

Donald Bruce Sharp

Michael Taher

A system including multiple processing devices such as application-specific integrated circuits (ASICs). Each processing device is provided with light at one or more wavelengths. Each processing device outputs data, at the wavelength of the optical signal received, to one or more interfaces. Each interface outputs a multiplexed signal containing data from each of the processing devices on a single optic fiber.

Systems and methods for optically connecting processing devices and interfaces

Devices, methods, and systems are provided. In one example, a device is described to include a bandwidth-constrained resource and a controller that dynamically allocates a proportional consumption of storage to the bandwidth-constrained resource thereby enabling the bandwidth-constrained resource to provide bandwidth to a consuming entity. The controller may allocate the proportional consumption of the storage to the bandwidth-constrained resource based on a current state of the bandwidth-constrained resource.

Dynamically reserved resource allocation

Methods, systems, and computer program products to modulate telemetry data as a function to represent the performance of a network and/or individual devices connected to the network. In embodiments, the method includes receiving telemetry data that has been sampled at a given point of time, wherein the telemetry data is associated with a performance metric of a device; processing the telemetry data as a function representing performance of the network device, wherein processing the telemetry data comprises modulating the telemetry data at the given point of time to previously sampled telemetry data based on the function; and demodulating the modulated telemetry data. In embodiments, the method also includes transferring the modulated telemetry data for reporting.

Method and system to modulate telemetry data

A network device includes a memory and a memory management circuit. The memory is to store a shared buffer. The memory management circuit is to estimate respective bandwidth measures for one or more queues used in processing packets in the network device, and to allocate and deallocate segments of the shared buffer to at least one of the queues based on the bandwidth measures.

Queue bandwidth estimation for management of shared buffers and allowing visibility of shared buffer status

A device, communication system, and method are provided. In one example, a system for routing traffic is described that includes a plurality of ports to facilitate communication over a network. The system also includes a controller to selectively activate or deactivate ports of the system based on queue depths and additional information to improve power efficiency of the system.

Adaptive port routing

Technologies for bi-directional encryption and decryption for underlay and overlay operations are described. One network device a path-selection circuit that operates in a first mode or a second mode. In the first mode, the path-selection circuit receives a first incoming packet on a first port, sends it to a security circuitry to decrypt the first incoming packet to obtain a first decrypted packet, sends the first decrypted packet to a processing circuitry to process the first decrypted packet to obtain a first outgoing packet, and sends the first outgoing packet to a second port of the network device. In the second mode, the path-selection circuit receives a second incoming packet on a third port, sends it to the processing circuitry to de-encapsulate the second incoming packet to obtain a second outgoing packet, and sends the second outgoing packet to a fourth port of the network device.

Bi-directional encryption/decryption device for underlay and overlay operations

A device includes ports, a packet processor, and a memory management circuit. The ports communicate packets over a network. The packet processor processes the packets using queues. The memory management circuit maintains a shared buffer in a memory and adaptively allocates memory resources from the shared buffer to the queues, maintains in the memory, in addition to the shared buffer, a shared-reserve memory pool for use by the queues, identifies, among the queues, a queue that requires additional memory resources, the queue having an occupancy that is (i) above a current value of a dynamic threshold, rendering the queue ineligible for additional allocation from the shared buffer, and (ii) no more than a defined margin above the current value of the dynamic threshold, rendering the queue eligible for allocation from the shared-reserve memory pool, and allocates memory resources to the identified queue from the shared-reserve memory pool.

Allocation of shared reserve memory

A parsing apparatus includes a packet-type identification circuit and a parser. The packet-type identification circuit is to receive a packet to be parsed, and to identify a packet type of the packet by extracting a packet-type identifier from a defined field in the packet. The parser is to store one or more parsing templates that specify parsing of one or more respective packet types. When the packet type of the packet corresponds to a parsing template among the stored parsing templates, the parser is to parse the packet in accordance with the stored parsing template. When the packet type of the packet does not correspond to any of the stored parsing templates, the parser is to parse the packet using an alternative parsing scheme.

Template-based packet parsing

An example method for dynamic packet routing using prioritized groups includes: receiving, by a processing device, routing information specifying a plurality of paths to a network destination, wherein each path is associated with a respective cost metric value; associating a first subset of the plurality of paths with a first priority routing group for the network destination, wherein each path of the first subset satisfies a first cost criterion based on a cost metric value of the path; associating a second subset of the plurality of paths with a second priority routing group for the network destination, wherein each path of the second subset satisfies a second cost criterion; and storing, in a forwarding information data structure, a first definition of the first priority routing group and a second definition of a second priority routing group.

Dynamic packet routing using prioritized groups

A method for filtering packets, the method including receiving a plurality of packets, each packet of the plurality of packets including a source address, and for each one packet of the plurality of packets, processing the one packet by performing the following: modifying the source address of the one packet to include at least an identifier of a flow to which the one packet belongs. Related apparatus and methods are also provided.

Stateful filtering systems and methods

Networking devices, systems, and methods are provided. In one example, a method includes receiving a packet at a networking device; evaluating the packet; based on the evaluation of the packet, truncating the packet from a first size to a second size that is smaller than the first size; and storing the truncated packet in a buffer prior to transmitting the truncated packet with the networking device.

Early and efficient packet truncation

A networking device and system are described, among other things. An illustrative system is disclosed to include a packet parser and a state machine that includes a NULL header state. The packet parser references the state machine to enter the NULL header state automatically in response to parsing a packet header of a predetermined type and then, while in the NULL header state, analyzes a subsequent set of bytes without advancing a parser pointer.

Parsing using null header node

Technologies for bi-directional encryption and decryption for underlay and overlay operations are described. One network device includes multiple ports, a network processing element, a programmable path-selection circuit, and a security IC. The programmable path-selection circuit is configured to operate in a first mode in which first outgoing packets are routed to the security integrated circuit to be encrypted before sending on one of the ports, and first incoming packets, received on one of the ports, are routed to the security integrated circuit to be decrypted. The programmable path-selection circuit is configured to operate in a second mode in which second incoming packets are routed to the security integrated circuit to be encrypted before processing by the network processing element and route second outgoing packets to the security integrated circuit to be decrypted after processing by the network processing element.

A network device includes one or more ports, a packet processor, and a memory management circuit. The one or more ports are to communicate packets over a network. The packet processor is to process the packets using a plurality of queues. The memory management circuit is to maintain a shared buffer in a memory and adaptively allocate memory resources from the shared buffer to the queues, to maintain in the memory, in addition to the shared buffer, a shared-reserve memory pool for use by a defined subset of the queues, to identify in the subset a queue that (i) requires additional memory resources, (ii) is not eligible for additional allocation from the shared buffer, and (iii) meets an eligibility condition for the shared-reserve memory pool, and to allocate memory resources to the identified queue from the shared-reserve memory pool.

Allocation of shared reserve memory to queues in a network device

Growing community of inventors

Campbell, CA, United States of America

Barak Gafni