Cisco Technology, Inc.

Amazon Technologies, Inc.

Cumulus Networks Inc.

Rivers, James Paul

James P Rivers

Scott A Emery

Stewart Findlater

Brian Arnold Petersen

David H Yen

Pauline Shuen

Gregory Lee DeJager

Navindra Yadav

Gnanaprakasam Pandian

Surendra Anubolu

Scott Andrew Emery

David R Oran

Silvano Gai

Ralph Droms

Claudio DeSanti

Colin John Whittaker

Harish Ramamurthy Devanagondi

Mark Stapp

John M Schnizlein

Sandeep Arvind Patel

Sean Convery

David Hsi-Chen Yen

Thomas James Edsall

Frederick Roland Schindler

Alan Michael Judge

Peter Clarke Newman

Balaji Satyanarayana Prabhakar

Alaa Adel Mahdi Hayder

Ke Zhang

William Frank Edwards

Linda T Cheng

Gordon Mackay

Bernard N Daines

Chaitanya Kodeboyina

Ravi Kumar Gadde

Wilson Kok

Andrew Richard Harding

Tahsin Erdogan

Murali Duvvury

David Talaski

Bradley D Erickson

Alan Yee

Poorya Saghari

Edward Nolan Leake

Eric Wiles

Shrijeet Subhabrata Mukherjee

Joshua Benjamin Levinson

Julien Ridoux

Leonid Livak

Paul Yu

Jeffrey D Provost

John K Chen

Michael Lawrence Regal

Mark Kamichoff

Yuan Yuan

John Oberstar

Rajesh Ekras Bawankule

Rita Ousterhout

Sean X Wang

Shrawan Chittoor Surender

Shriram Srinivasan

Abhishek Vinayak Naik

Chinchu Merine Joseph

Michael Regal

A clock signal in a clock distribution network is transmitted using network packets with the clock signal embedded within a bit of the network packets. The network packets can include a preamble used to create phase alignment between a timing pulse and a bit position of a Start of Frame Delimiter (SFD) within the packet. The clock signal associated with the packet occurs when the SFD is detected. In one example, the SFD is detected when two consecutive bits of equal value are received and the timing of the clock signal is such that the clock signal occurs when the second consecutive bit is received. By including a clock signal within a packet, additional information can be transmitted with the clock signal. For example, authentication and validation information can be included, a time stamp, a message type, a frame check sequence, a clock status, a number of hops from the root, etc.

Transmitting clock signals in network packets

A host-to-ToR switch advertisement protocol can be used at a data center network edge to receive IP advertisements from hosts and convey ToR switch availability status (e.g., in service, intentionally out of service, or unintentionally out of service) back to hosts and vice versa. The host server computers can have the ability to advertise an arbitrary number of IPv4 or IPv6 prefixes to the network through the ToR switches. A single host server computer can leverage multiple instances across multiple ports to the same switch or different switches to implement load balancing (e.g., equal cost multipath (ECMP)) and fault tolerance. The host server computers benefit from an increased availability by receiving network traffic via multiple ToR switches, a single ToR switch with multiple links, or a combination of both.

Network rack of host server computers with increased network availability

A clock signal in a clock distribution network is transmitted using network packets with the clock signal embedded within a bit of the network packets. The clock signal is adjusted to account for propagation delay in transmitting the clock signal throughout a clock distribution network. The propagation delay is computed using a round-trip packet. When a previous clock signal is received, a timer is set to a local clock's estimate of when the next clock signal will occur minus the propagation delay to the downstream device. When this timer expires, the clock signal is sent to the downstream device, which will allow it to arrive at the downstream device at the same time as the next clock signal is received on the local device.

Method and apparatus for accounting for propogation delay with distributing a clock signal

A clock selection circuit allows seamless switching between different clock signals in a clock distribution network. The clock selection circuit can be an Integrated Circuit (IC). The clock signals can be analyzed by a processor in communication with the IC to ensure the clock signals are validated. Analysis can include comparing time stamps between received pulses of the clock signals to determine if the clock signals are occurring at regular intervals. The processor can then assign a priority order to the clock signals and select one of the clock signals to use. An identifier associated with the selected clock signal can be programmed into the IC. The IC can then redistribute the selected clock signal to multiple other ICs in a hierarchical clock distribution network. Ultimately, the distributed clock signal can be received by server computers to ensure instances being executed have accurate and synchronized timing.

Clock selection in a clock distribution network

Systems and methods are provided for highly accurate synchronization of machine instances in a distributed, hosted computing environment to a reference timekeeper. In addition to a general communication network accessible to machine instances, the distributed environment includes a second network dedicated to carrying time information, such as a pulse-per-second (PPS) signal to isolated timing hardware within host computing devices. The isolated timing hardware can use the PPS signal, along with a reference time, to set a hardware clock. The isolated timing hardware can further provide an interface to machine instances that enables the instances to read the time of the hardware clock. This configuration enables many instances can share access to a single reference timekeeper, thus synchronizing those instances to a much higher accuracy than in traditional network-based time protocols.

Virtualized hardware clocks for providing highly accurate time information in hosted machine instances

A cage assembly may define a part of a computing appliance. The assembly can include a cage formed by walls and defining a continuous interior volume bounded by a top face, a bottom face, a right face, a left face, a front face, and a rear face. A first edge connector may be within a leftward portion of the interior volume. A second edge connector may be within a rightward portion of the interior volume and positioned laterally from the first edge connector. The arrangement may permit engagement with multiple individual-width plugs (e.g., each bearing an individual paddle card) or a single plug (e.g., bearing multiple paddle cards spaced laterally from one another and aligned for engagement with the first and second edge connectors within the cage).

Multiple-width plug arrangement for networking connectivity

To accommodate graceful offloading of connections from a port of a network element, thereby allowing the port to be decommissioned, both existing and new connections allocated to that port according to a load balancing protocol are replicated on a backup port. Thereafter, or concurrently therewith, the port is configured to drop acknowledgements of the new connections, and the port is monitored so as to eventually identify an absence of connections thereon. At that time, the port may be decommissioned inasmuch as all new connections will have been established on the backup port and no previously existing connections on the original port will remain.

Flow based session drain director

In one embodiment, a subnet-scoped multicast packet is received on an interface of a forwarding device that is connected to a host device of a subnet of a forwarding domain. The received subnet-scoped multicast packet is transmitted from one or more other interfaces of the forwarding device that are connected to one or more other host devices of the subnet. The received subnet-scoped multicast packet is also encapsulated with an additional header. The encapsulated subnet-scoped multicast packet is forwarded from the forwarding device to an intermediate router which routes the encapsulated subnet-scoped multicast packet to one or more other forwarding devices configured to decapsulate the encapsulated subnet-scoped multicast packet and transmit the decapsulated subnet-scoped multicast packet to one or more connected host devices of an additional portion of the subnet.

Subnet scoped multicast/broadcast packet distribution over a routed network

Methods and apparatus are disclosed for applying network policy to communications originating at operating system virtual interfaces. In an example embodiment, a network device is networked with a switch. The network device may include a first operating system interface, a virtualization adapter, and an input output port. In an example embodiment, the virtualization adapter receives a first frame from the first operating system interface. The virtualization adapter may tag the first frame to indicate an association between the first frame and the first operating system interface. The first frame may then be transmitted with a second frame being associated with a second operating system interface, to the switch via the input output port. In an example embodiment, the switch is configured to receive the frame, examine a tag and then to enforce a network policy upon the first frame, based on the tag.

Apparatus and method for applying network policy at virtual interfaces

This document discusses, among other things, applying network policy at a network device. In an example embodiment fiber channel hard zoning information may be received that indicates whether a fiber channel frame is permitted to be communicated between two fiber channel ports. Some example embodiments include identifying a media access control addresses associated with the fiber channel ports. An example embodiment may include generating one or more access control entries based on the fiber channel identifications of the fiber channel ports and the zoning information. The access control entries may be distributes to an Ethernet port to be inserted into an existing access control list and used to enforce a zoning policy upon fiber channel over Ethernet frames.

Apparatus and method for applying network policy at a network device

Methods and systems for synchronizing network configuration state tables between an operating system kernel and an attached ASIC switch device are disclosed. An instruction to update a target network configuration state table in either an operating system kernel or an attached ASIC switch device is received. Data from the target network configuration state table and data from a corresponding network configuration state table are retrieved. Values of data in the two tables are compared and an instruction is sent to update the target network configuration state table with one or more values from the corresponding network configuration state table.

Systems and methods for accelerating networking functionality

In one embodiment, an apparatus may include a first interface configured to be communicatively coupled, via a network, to a second interface and a fiber channel services module. The first interface may be configured to receive a fiber channel service from the fiber channel services module, establish communication with the second interface, and communicate a fiber-channel-over-Ethernet (FCoE) frame to the second interface, via a forwarder that forwards the FCoE frame without employing a fiber channel switching element. Other embodiments are described and claimed.

Providing fibre channel services and forwarding fibre channel over ethernet frames

Subnet scoped multicast/broadcast packet distribution mechanism over a routed network

A method and system to manage network traffic congestion in networks with link layer flow control is provided. The system comprises a physical queue monitor configured to monitor a state of a physical queue at a network device, a link layer flow control activator configured to activate link layer flow control based on the state of the physical queue, a proxy queue control module, a proxy queue monitor to monitor the state of the proxy queue, and a transport layer flow control activator. The proxy queue control module may be configured to update a state of a proxy queue based on the state of the physical queue. The proxy queue monitor may be configured to monitor the state of the proxy queue. The transport layer flow control activator may be configured to activate transport layer flow control based on the state of the proxy queue.

Method and system to manage network traffic congestion in networks with link layer flow control

This document discusses, among other things, applying network policy at a network device. In an example embodiment fiber channel hard zoning information may be received that indicates whether a fiber channel frame is permitted to be communicated between two fiber channel ports. Some example embodiments include identifying a media access control address associated with the fiber channel ports. An example embodiment may include generating one or more access control entries based on the fiber channel identifications of the fiber channel ports and the zoning information. The access control entries may be distributed to an Ethernet port to be inserted into an existing access control list and used to enforce a zoning policy upon fiber channel over Ethernet frames.

Growing community of inventors

Saratoga, CA, United States of America

James P Rivers