Intel Corporation

Albert S Cheng

Thomas D Lovett

Michael A Parker

Todd M Rimmer

Mike Parker

Steve Scott

Robert L Alverson

Steven F Hoover

Francesc Guim Bernat

Mark S Birrittella

James Kunz

Gaspar Mora Porta

Roberto Penaranda Cebrian

Gregory J Hubbard

John J Kim

Scott S Diesing

Nan Ni

Michael S Parker

James A Kunz

John Kim

This disclosure describes systems, devices, methods and computer readable media for enhanced network communication for use in higher performance applications including storage, high performance computing (HPC) and Ethernet-based fabric interconnects. In some embodiments, a network controller may include a transmitter circuit configured to transmit packets on a plurality of virtual lanes (VLs), the VLs associated with a defined VL priority and an allocated share of network bandwidth. The network controller may also include a bandwidth monitor module configured to measure bandwidth consumed by the packets and an arbiter module configured to adjust the VL priority based on a comparison of the measured bandwidth to the allocated share of network bandwidth. The transmitter circuit may be further configured to transmit the packets based on the adjusted VL priority.

Traffic class arbitration based on priority and bandwidth allocation

In some embodiments, the invention involves using a weighted arbiter switch to provide fairness in passing input streams through a plurality of input ports to an output port. The weighted arbiter switches may be combined in a hierarchical architecture to enable routing through many levels of switches. Each input port has an associated flow counter to count input stream traffic through the input port. An arbiter switch uses the flow counts and weights from arbiter switches at a lower level in the hierarchy to generate a fairly distributed routing of input streams through the output port. Other embodiments are described and claimed.

System and method to enable fairness on multi-level arbitrations for switch architectures

Technologies for determining a root of congestion include a network switch. The network switch is to operate arbiter units in at least one upstream stage at a packet transfer rate that is greater than a packet transfer rate of an arbiter unit in an output stage, determine whether an input buffer of a remote network switch in communication with the output stage has sustained congestion over a first predefined time period, determine whether an output buffer of the arbiter unit in the output stage has sustained congestion over a second predefined time period, and determine, as a function of whether the input buffer of the remote network switch has sustained congestion and whether the output buffer of the arbiter unit in the output stage has sustained congestion, whether the network switch is a root of congestion.

Technologies for efficiently determining a root of congestion with a multi-stage network switch

Technologies for balancing throughput across input ports include a network switch. The network switch is to generate, for an arbiter unit in a first stage of a hierarchy of stages of arbiter units, turn data indicative of a set of turns in which to transfer packet data from devices connected to input ports of the arbiter unit. The network switch is also to transfer, with the arbiter unit, the packet data from the devices in the set of turns. Additionally, the network switch is to determine weight data indicative of the number of turns represented in the set and provide the weight data from the arbiter unit in the first stage to another arbiter unit in a subsequent stage to cause the arbiter unit in the subsequent stage to allocate a number of turns for the transfer of the packet data from the arbiter unit in the first stage.

Technologies for balancing throughput across input ports of a multi-stage network switch

Apparatuses, methods and storage medium associated with routing data in a switch are provided. In embodiments, the switch may include route lookup circuitry determine a first set of output ports that are available to send a data packet to a destination node. The lookup circuitry may further select, based on respective congestion levels associated with the first set of output ports, a plurality of output ports for a second set of output ports from the first set of output ports. An input queue of the switch may buffer the data packet and route information associated with the second set of output ports. The switch may further include route selection circuitry to select a destination output port from the second set of output ports, based on updated congestion levels associated with the output ports of the second set of output ports. Other embodiments may be described and/or claimed.

Apparatus and method for routing data in a switch

Apparatuses and methods associated with distributing congestion information in a switch are provided. In embodiments, the switch includes a plurality of ports arranged in a plurality of rows and a plurality of columns. The switch further includes a plurality of daisy-chain buses, individual daisy-chain buses coupling the ports of a respective row to one another in a daisy-chain. The switch further includes a plurality of column buses, individual column buses coupling an individual port of the plurality of ports to the other ports of the respective column. Individual ports of a respective row receive congestion information from the other ports of the row via the respective daisy-chain bus and pass the congestion information to the other ports of the respective column via the respective column bus. Other embodiments may be described and/or claimed.

Apparatus and method for distribution of congestion information in a switch

Generally, this disclosure provides systems, methods and computer readable media for switching systems employing distributed routing tables to provide improved support for multiple network topologies, configurations and routing modes. A routing system of a switch, for routing a packet from an input port of a switch to an output port, may include memory modules to store routing tables configured to store entries, each entry configured to store output port identifiers. The routing system may also include a table address generation module configured to select an entry from the routing tables based in part on a first subset of a Destination Location ID (DLID) associated with the packet. The table address generation module may further be configured to select one or more output ports from the selected entries, the selected output ports based in part on a second subset of the DLID.

Distributed routing table system with improved support for multiple network topologies

A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.

Table-driven routing in a dragonfly processor interconnect network

Apparatuses, methods and storage medium associated with buffering data in a switch are provided. In embodiments, the switch may include a plurality of queue buffers, a plurality of queues respectively associated with the plurality of queue buffers, a shared buffer, and a queue point controller coupled with the plurality of queue buffers and the shared buffer. In embodiments the queue point controller may be configured to determine an amount of available space in a selected queue buffer of the plurality of queue buffers. The queue point controller may be further configured to allocate at least a portion of the shared buffer to a selected queue that is associated with the selected queue buffer. In embodiments, this allocation may be based on the amount of available space determined in the selected queue buffer. Other embodiments may be described and/or claimed.

Apparatus and method for buffering data in a switch

Technologies for scalable local addressing include one or more managed network devices coupled to one or more computing nodes via high-speed fabric links. A computing node may transmit a data packet including a destination local identifier (DLID) that identifies the destination computing node. The DLID may be 32, 24, 20, or 16 bits wide. The managed network device may determine whether the DLID is within a configurable multicast address space and, if so, forward the data packet to a multicast group. The managed network device may also determine whether the DLID is within a configurable collective address space and, if so, perform a collective acceleration operation. The number of top-most bits set in a multicast mask and the number of additional top-most bits set in a collective mask may be configured. Multicast LIDs may be converted between different bit lengths. Other embodiments are described and claimed.

Technologies for scalable local addressing in high-performance network fabrics

Apparatuses, methods and storage media associated with multiple multi-drop buses in a switch are provided herein. In some embodiments, the switch may include a multi-drop row bus to transmit a plurality of frames in a row dimension of the matrix switch and a multi-drop column bus to transmit the plurality of frames in a column dimension of the matrix switch. The switch may further include an input port to receive the plurality of frames and an output port to output the plurality of frames from the matrix switch. Other embodiments may be described and/or claimed.

Apparatus and method for using multiple multi-drop buses

Methods, apparatus, and systems for implementing hierarchical and lossless packet preemption and interleaving to reduce latency jitter in flow-controller packet-based networks. Fabric packets are divided into a plurality of data units, with data units for different fabric packets buffered in separate buffers. Data units are pulled from the buffers and added to a transmit stream in which groups of data units are interleaved. Upon receipt by a receiver, the groups of data units are separated out and buffered in separate buffers under which data units for the same fabric packets are grouped together. In one aspect, each buffer is associated with a respective virtual lane (VL), and the fabric packets are effectively transferred over fabric links using virtual lanes. VLs may have different levels of priority under which data units for fabric packets in higher-priority VLs may preempt fabric packets in lower-priority VLs. By transferring data units rather than entire packets, transmission of a packet can be temporarily paused in favor of a higher-priority packet. Multiple levels of preemption and interleaving in a nested manner are supported.

Hierarchical/lossless packet preemption to reduce latency jitter in flow-controlled packet-based networks

Method, apparatus, and systems for implementing flexible credit exchange within high performance fabrics. Available buffer space in a receive buffer on a receive-side of a link is managed and tracked at the transmit-side of the link using credits. Peer link interfaces coupled via a link are provided with receive buffer configuration information that specifies how the receive buffer space in each peer is partitioned and space allocated for each buffer, including a plurality of virtual lane (VL) buffers. Credits are used for tracking buffer space consumption and in credits are returned from the receive-side indicating freed buffer space. The peer link interfaces exchange credit organization information to inform the other peer of how much space each credit represents. In connection with data transfer over the link, the transmit-side de-allocates credits based on an amount of buffer space to be consumed in applicable buffers in the receive buffer. Upon space being freed in the receive buffer, the receive-side returns credit ACKnowledgements (ACKs) identifying a VL for which space has been freed.

Growing community of inventors

Bellevue, WA, United States of America

Albert S Cheng