Cisco Technology, Inc.

Sun Microsystems, Inc.

Meta Platforms, Inc.

Linda T Cheng

Denton E Gentry, Jr

Ganesh Srinivasa Bhat

Subrat Mohanty

Ali Ahmad Tareen

Manpreet Singh Sandhu

Shimon Muller

John E Watkins

James P Rivers

Abdulkadir Utku Diril

Olivia Wu

Stewart Findlater

Scott A Emery

Pankaj Kansal

Jimit Maheta

Methods and systems are disclosed. The method comprises: designating a first plurality of links from a first stack segment to a second stack segment as a first etherchannel link; designating a second plurality of links from the first stack segment to a third stack segment as a second etherchannel link, where the second stack segment and the third stack segment are in communication with a fourth stack segment; designating the first etherchannel link and the second etherchannel link as members of a hierarchical etherchannel link; and sending a packet from the first stack segment to the fourth stack segment using the hierarchical etherchannel link.

Method and apparatus to optimize multi-destination traffic over etherchannel in stackwise virtual topology

A system including an array of processing elements, a plurality of periphery crossbars and a plurality of storage components is described. The array of processing elements is interconnected in a grid via a network on an integrated circuit. The periphery crossbars are connected to a plurality of edges of the array of processing elements. The storage components are connected to the periphery crossbars.

Optimizing NOC performance using crossbars

Methods and systems for configuring a first switch for frontside networking. The method includes: discovering the capabilities of a second switch connected to the first switch, where the capabilities include frontside stacking; changing a port on the first switch connected to the second switch from a standard port to a frontside stacking port; discovering a topology of a network containing the first switch and the second switch; and creating a drop table having entries for at least the first switch and the second switch.

Systems and methods for enabling frontside stacking of switches

A cross stack port aggregation method and system associates a destination index with a received packet when it is provided to devices in the stack. Each device utilizes the destination index to access a descriptor identifying ports in the device included in the port aggregation group. An index generated from packet address data is used to select a bit in a group mask unique to each port.

Defining link aggregation across a stack

A high performance network interface receives network traffic in the form of packets. Prior to being transferred to a host computer, a packet is stored in a packet queue. A system and method are provided for randomly discarding a packet if the rate of packet transfers cannot keep pace with the rate of packet arrivals at the queue. When a packet must be dropped a selected packet may be discarded as it arrives at the queue, or a packet already in the queue may be dropped. A packet queue is apportioned into multiple regions, any of which may overlap or share a common boundary. A probability indicator is associated with a region to specify the probability of a packet being discarded when the level of traffic stored in the queue is within the region. A counter may be employed in conjunction with a probability indicator to identify individual packets. Probability indicators may differ from region to region so that the probability of discarding a packet fluctuates as the level of traffic stored in the queue changes. In addition to selecting packets to be dropped on a random basis, information gleaned from a packet may be applied to prevent certain types of packets from being dropped. The information derived from a packet may be obtained during a procedure in which one or more of the packet's headers are parsed. By parsing a packet, it may be determined whether the packet conforms to a pre-selected protocol.

Method and apparatus for early random discard of packets

A system and method are provided for transferring a packet received from a network to a host computer according to an operation code associated with the packet. A packet received at a network interface is parsed to retrieve information from a header portion of the packet. A flow key is generated for a received packet that was formatted with one of a set of predetermined protocols. A packet's flow key identifies a communication flow that comprises the packet. Based on some of the retrieved information, a code is associated with the packet to inform a transfer engine how the packet should be transferred to host memory. Based on a packet's code, the transfer engine stores the packet in one or more host memory buffers. If the packet was formatted with one of the set of predetermined protocols, its data is re-assembled in a re-assembly buffer with data from other packets in the same communication flow. Re-assembled data may be provided to a destination application or user through page flipping. If the packet is being re-assembled, a header portion of the packet is stored in a separate header buffer. If the packet is not being re-assembled, it is stored in its entirety in the header buffer if it is smaller than a predetermined threshold. If a non-re-assembled packet is larger than the threshold for the header buffer, it is stored in another type of buffer for larger non-re-assembled packets. After a packet is stored in a buffer, the transfer engine informs the host computer of the packet by configuring a descriptor with information on the packet and releasing the descriptor to the host computer.

Method and apparatus for data re-assembly with a high performance network interface

A network interface is polled by a process operating on a host computer system. Each time the network interface is polled the process determines whether any packets have been transferred to the host. If so, they are processed. Interrupts that would normally be issued from the network interface in response to the transfer of packets are suppressed or postponed during the polling mode of operation. If, however, a predetermined period of time has elapsed or a predetermined number of packets have been transferred since a previous poll or a previous interrupt, then an interrupt may be generated.

Method and apparatus for indicating an interrupt in a network interface

A high performance network interface is provided for receiving a packet from a network and transferring it to a host computer system. A header portion of a received packet is parsed by a parser module to determine the packet's compatibility with, or conformance to, one or more pre-selected protocols. If compatible, a number of processing functions may be performed to increase the efficiency with which the packet is handled. In one function, a re-assembly engine re-assembles, in a re-assembly buffer, data portions of multiple packets in a single communication flow or connection. Header portions of such packets are stored in a header buffer. An incompatible packet may be stored in another buffer. In another function, a packet batching module determines when multiple packets in one flow are transferred to the host computer system, so that their header portions are processed collectively rather than being interspersed with headers of other flows' packets. In yet another function, the processing of packets through their protocol stacks is distributed among multiple processors by a load distributor, based on their communication flows. A flow database is maintained by a flow database manager to reflect the creation, termination and activity of flows. A packet queue stores packets to await transfer to the host computer system, and a control queue stores information concerning the waiting packets. If the packet queue becomes saturated with packets, a random packet may be discarded. An interrupt modulator may modulate the rate at which interrupts associated with packet arrival events are issued to the host computer system.

High performance network interface

An arbiter implements an arbitration scheme that allows a requestor that has been granted use of a resource to use the resource for a predetermined number of data transfers before relinquishing the resource. At the time a requestor is granted use of the resource, a register is loaded with a value that defines a limit on the how much the requestor can use the resource. More particularly, the value defines a number of bytes that the requester is allowed to transfer before being required to relinquish use of the resource. The value is decremented each time the requestor makes a data transfer, as long as there is also a pending request from another requestor. If there is no pending request from another requester, the value is not decremented even though the requestor has transferred data. Until the time that the value is decremented to zero, the requester may repeatedly make requests to the arbiter for use of the resource, and the arbiter will continue to grant them. Upon the value becoming zero, however, the arbiter reverts to its original state of handling incoming requests for the resource.

Growing community of inventors

San Jose, CA, United States of America

Linda T Cheng