Cisco Technology, Inc.

Pensando Systems Inc.

Cisco Systems Inc.

David Systems, Inc.

Crescendo Communications, Inc.

Mario Mazzola

Luca Cafiero

Satyanarayana Nishtala

Philip R Manela

Silvano Gai

Michael Brian Galles

Krishna Doddapaneni

Sarat Kamisetty

Tom Edsall

Francis Matus

Massimo Prati

Peter Clarke Newman

Georges Akis

Dmitry Barsky

Maurilio DeNicolo

Philip Manela

PCIe devices installed in host computers communicating with service nodes can provide virtualized NVMe over fabric services. A workload on the host computer can submit an SQE on a NVMe SQ. The PCI device can read the SQE to obtain a command identifier, an OpCode, and a namespace identifier (NSID). The SQE can be used to produce a LTP packet that includes the opcode, the NSID, and a request identifier. The LTP packet can be sent to the service node, which may access a SAN in accordance with the opcode and NSID, and can respond to the LTP with a second LTP that includes the request identifier and a status indicator. The PCI device can use the status indicator and the request identifier to produce a CQE that is placed on a NVMe CQ associated with the SQ.

Methods and systems for providing a virtualized NVMe over fabric service

PCIe devices installed in host computers communicating with service nodes can provide virtualized and high availability PCIe functions to host computer workloads. The PCIe device can receive a PCIe TLP encapsulated in a PCIe DLLP via a PCIe bus. The TLP includes a TLP address value, a TLP requester identifier, and a TLP type. The PCIe device can terminate the PCIe transaction by sending a DLLP ACK message to the host computer in response to receiving the TLP. The TLP packet can be used to create a workload request capsule that includes a request type indicator, an address offset, and a workload request identifier. A workload request packet that includes the workload request capsule can be sent to a virtualized service endpoint. The service node, implementing the virtualized service endpoint, receives a workload response packet that includes the workload request identifier and a workload response payload.

Methods and systems for loosely coupled PCIe service proxy over an IP network

This document discusses, among other things, an example system and methods for memory expansion. An example embodiment includes receiving a memory request from a memory controller over a channel. Based on the memory request, the example embodiment includes selecting a location in memory to couple to a sub-channel of the channel and configuring the set of field effect transistors to couple the channel with the sub-channel. In the example embodiment, data may be allowed to flow between the memory controller and the location in the memory over the channel and the sub-channel.

System and methods for memory expansion

A multi-stage network switch comprises a plurality of ingress port subsystems each comprising one or more ingress ports configured to receive packets. The switch also comprises a plurality of unscheduled crossbar switching elements connected to the ingress port subsystems that are configured to receive one or more packets from at least one of the ingress port subsystems. The switch further comprises a plurality of egress port subsystems each comprising a memory and a plurality of egress ports. The memory comprises at least one shared egress buffer configured to receive any packets forwarded by the crossbar switching elements from the ingress port subsystems directed to the egress port subsystem, and the egress ports are configured to transmit the packets received in the shared egress buffer.

Shared egress buffer in a multi-stage switch

This document discusses, among other things, an example system and methods for memory expansion. An example embodiment includes receiving first initialization data from a physical dual inline memory module (DIMM) and converting the first initialization data to second initialization data of a logical DIMM mapped to the physical DIMM. The example embodiment may further include programming a memory controller based on the second initialization data.

This document discusses, among other things, an example system and methods for memory expansion. An example embodiment includes detecting a memory command directed to a logical rand and a number of physical ranks mapped to the logical rank. The example embodiment may also include issuing the memory command to the number of physical ranks based on determining that the memory command is to be issued to the number of physical ranks.

A switching bus architecture enables efficient transfer of data within a network switch having a plurality of ports interconnected by a high-performance switching bus. The architecture is preferably implemented as novel port interface and forwarding engine circuitry that cooperate to efficiently transmit data to, and receive data from, the switching bus in accordance with a 2-tier arbitration policy that ensures adequate port access to the bus. As a result of such a cooperating arrangement, the architecture improves the transfer efficiency of the switch by providing all ports sufficient bus access to convey accurate data throughout the switch.

Architecture for an expandable transaction-based switching bus

An address translation mechanism quickly and efficiently renders forwarding decisions for data flames transported among ports of a high-performance switch on the basis of, inter alia, virtual local area network (VLAN) associations among the ports. The translation mechanism comprises a plurality of forwarding tables, each of which contains entries having unique index values that translate to selection signals for ports destined to received the data frames. Each port is associated with a unique index value and a VLAN identifier to facilitate multicast data transfers within the switch at accelerated speeds and addressing capabilities.

Address translation mechanism for a high-performance network switch

A multilevel encoding scheme for transmitted data that encodes data in a multilevel code wherein the amplitude of any transition is always exactly one level during any time interval. A single-level transition between any two adjacent levels during a time interval represents a logical '1'; no transition during a time interval represents a logical '0'. In a specific embodiment, modulation is limited to three defined amplitude levels equally space in amplitude and encoding is according to a three-level code. A four-bit to five-bit encoding scheme may be used to distribute bits for minimizing d.c. offset. The input data is preferably further scrambled to minimize aberrations in the emissions spectrum of signal carried over unshielded media.

Method and apparatus for multilevel encoding for a local area network

A high-speed digital transceiver is provided for use in a PBX environment comprising twisted-pair wire cables interconnecting like transceivers, each transceiver being operative to exchange voice, data and control information in a packetized format over a common twisted-pair cable. Specifically, each transceiver communicates packetized pulse code modulated information in pure Alternate Mark Inverted (AMI) coding, that is, without the introduction of bipolar violation pulses to provide timing. Frame synchronization is acquired on the first pulse by the use of a digital circuit deriving synchronization from a local high-speed clock. The use of a high-speed clock-driven digital circuit for synchronization acquisition eliminates the need for a phase-locked loop synchronization scheme and its concomitant finite acquisition delay. In addition, a receiving section employs a threshold selecting circuit which switches or makes thresholds in response to an expectation of the absence any bipolar violation in the transmitted signal. The effect of intersymbol interference are further minimized by provision of digital precompensation in the transmitted signal to maximize the slew rate between consecutive pulses. The precompensation scheme is based on a knowledge of the bit pattern and the amount of energy contained in a sequence of bits.

Growing community of inventors

San Jose, CA, United States of America

Mario Mazzola