Mellanox Technologies, Ltd.

Shahaf Shuler

Noam Bloch

Ariel Shahar

Dotan David Levi

Liran Liss

Ilan Pardo

Gil Bloch

Daniel Marcovitch

Natan Manevich

Ariel Almog

Yossef Itigin

Oren Duer

Eliav Bar-Ilan

Eitan Hirshberg

Ran Avraham Koren

Sayantan Sur

Stephen Anthony Bernard Jones

George Elias

Lion Levi

Richard Leigh Graham

Miriam Menes

Peter Paneah

Uria Basher

Ofer Hayut

Ahmad Atamli

Chen Rozenbaum

Tzuriel Katoa

Gabi Liron

Gary Mataev

Nizan Atias

Adi Maymon

Shaul Arazi

Meni Orenbach

A kernel comprising at least one dynamically configurable parameter is submitted by a processor. The kernel is to be executed at a later time. Data is received after the kernel has been submitted. The at least one dynamically configurable parameter of the kernel is updated based on the data. The kernel having the at least one updated dynamically configurable parameter is executed after the at least one dynamically configurable parameter has been updated.

Kernel optimization and delayed execution

A networking device, system, and method of operating a networking device are provided. The illustrative networking device is disclosed to include one or more physical ports, an emulated switch positioned between the one or more physical ports and a host device, and one or more emulated devices positioned between the emulated switch and the one or more physical ports. The one or more emulated devices may be configured to populate the one or more physical ports.

Support for multiple hot pluggable devices via emulated switch

The technology disclosed herein enables selective clearing of memory regions upon a context switch. An example method includes the operations of: receiving a memory access request referencing a memory region; determining an identifier of a current execution context associated with the memory region; determining an identifier of a previous execution context specified by metadata associated with the memory region; responsive to determining that the identifier of the current execution context does not match the identifier of the previous execution context, updating the metadata associated with the memory region to store the identifier of the current execution context; clearing at least a part of the memory region; and processing the memory access request with respect to the memory region.

Maintaining data confidentiality in shared computing environments

A network adapter includes a network interface, a bus interface, a hardware-implemented data-path and a programmable Data-Plane Accelerator (DPA). The network interface is to communicate with a network. The bus interface is to communicate with an external device over a peripheral bus. The hardware-implemented data-path includes a plurality of packet-processing engines to process data units exchanged between the network and the external device. The DPA is to expose on the peripheral bus a User-Defined Peripheral-bus Device (UDPD), to run user-programmable logic that implements the UDPD, and to process transactions issued from the external device to the UDPD by reusing one or more of the packet-processing engines of the data-path.

Programmable user-defined peripheral-bus device implementation using data-plane accelerator (DPA)

In one embodiment, a data communication device includes a network interface controller to process packets received from at least one of a host device for sending over a network, and at least one remote device over the network, at least one processor to execute computer instructions to receive a configuration, and extract filtering rules from the configuration, and at least one hardware accelerator to receive the filtering rules from the at least one processor, and filter the packets based on the rules so that some of the packets are dropped and some of the packets are forwarded to the at least one processor to send data based on the forwarded packets to another device.

Efficient and flexible flow inspector

A communication system includes at least one send queue, containing send queue entries pointing to packets to be transmitted over a network by packet sending circuitry. A clock work queue contains clock queue entries to synchronize sending times of the packets pointed to by the send queue entries. At least one arming queue contains arming queue entries to arm the clock work queue at selected time intervals.

Clock queue with arming and/or self-arming features

A timing system including timing circuitry which includes an arming queue, a clock work queue, and a clock completion queue. At least the clock work queue is to provide timing information, and the arming queue is to arm the clock work queue. Related apparatus and methods are also provided.

An apparatus includes a processor, configured to designate a memory region in a memory, and to issue (i) memory-access commands for accessing the memory and (ii) a conditional-fence command associated with the designated memory region. Memory-Access Control Circuitry (MACC) is configured, in response to identifying the conditional-fence command, to allow execution of the memory-access commands that access addresses within the designated memory region, and to defer the execution of the memory-access commands that access addresses outside the designated memory region, until completion of all the memory-access commands that were issued before the conditional-fence command.

Processor with conditional-fence commands excluding designated memory regions

A method for communication includes receiving in a computer system a request from a peer computer system. Upon finding that the computer system is currently not ready to process the request, a Negative Acknowledgement (NAK) message is sent from the computer system to the peer computer system, at a sending time that is derived from a time at which the computer system is ready to process the request.

Delayed responder-not-ready negative acknowledgement

In a fabric of network elements one network element has an object pool to be accessed stored in its memory. A request for atomic access to the object pool by another network element is carried out by transmitting the request through the fabric to the one network element, performing a remote direct memory access to a designated member of the object pool, atomically executing the request, and returning a result of the execution of the request through the fabric to the other network element.

Atomic access to object pool over RDMA transport network

A network adapter includes a network interface and circuitry. The network interface is assigned a single network address in a communication network, and is configured to receive, from one or more other nodes over the communication network, messages that are destined for processing by multiple threads in one or more processing cores of a network node including the network adapter, but are nevertheless addressed to the single network address. The circuitry is configured to hold a distribution rule for distributing the messages among multiple Receive Queues (RQs) that are accessible by the threads, and to select for each message received via the network interface a respective RQ, by applying the distribution rule to the message.

Distribution of messages to queues in a distributed computing environment

An MPI collective operation carried out in a fabric of network elements by transmitting MPI messages from all the initiator processes in an initiator node to designated ones of the responder processes in respective responder nodes. Respective payloads of the MPI messages are combined in a network interface device of the initiator node to form an aggregated MPI message. The aggregated MPI message is transmitted through the fabric to network interface devices of responder nodes, disaggregating the aggregated MPI message into individual messages, and distributing the individual messages to the designated responder node processes.

In-node aggregation and disaggregation of MPI alltoall and alltoallv collectives

A method for communication includes posting, by a software process, a set of buffers in a memory of a host processor and creating in the memory a list of labels associated respectively with the buffers. The software process pushes a first part of the list to a network interface controller (NIC), while retaining a second part of the list in the memory under control of the software process. Upon receiving a message containing a label, sent over a network, the NIC compares the label to the labels in the first part of the list and, upon finding a match to the label, writes data conveyed by the message to a buffer in the memory. Upon a failure to find the match in the first part of the list, the NIC passes the message from the NIC to the software process for handling using the second part of the list.

Growing community of inventors

Karkom, Israel

Shahaf Shuler