Google Inc.

Robert Edman Felderman

Peter H Hochschild

Alexander Lloyd

Wilson Cheng-Yi Hsieh

Michael James Boyer Epstein

Abdul Kabbani

Dennis Charles Abts

Guenter E Roeck

Hain-Ching Liu

Roy Michael Bannon

The present technology proposes techniques for generating globally coherent timestamps. This technology may allow distributed systems to causally order transactions without incurring various types of communication delays inherent in explicit synchronization. By globally deploying a number of time masters that are based on various types of time references, the time masters may serve as primary time references. Through an interactive interface, the techniques may track, calculate and record data relative to each time master thus providing the distributed systems with causal timestamps.

Generating globally coherent timestamps

Aspects and implementations of the present disclosure are directed to an indirect generalized hypercube network in a computer network facility. Servers in the computer network facility participate in both an over-subscribed fat tree network hierarchy culminating in a gateway connection to external networks and in an indirect hypercube network interconnecting a plurality of servers in the fat tree. The participant servers have multiple network interface ports, including at least one port for a link to an edge layer network device of the fat tree and at least one port for a link to a peer server in the indirect hypercube network. Servers are grouped by edge layer network device to form virtual switches in the indirect hypercube network and data packets are routed between servers using routes through the virtual switches. Routes leverage properties of the hypercube topology. Participant servers function as destination points and as virtual interfaces for the virtual switches.

Systems and methods for routing data through data centers using an indirect generalized hypercube network

Aspects and implementations of the present disclosure are directed to an indirect generalized hypercube network in a data center. Servers in the data center participate in both an over-subscribed fat tree network hierarchy culminating in a gateway connection to external networks and in an indirect hypercube network interconnecting a plurality of servers in the fat tree. The participant servers have multiple network interface ports, including at least one port for a link to an edge layer network device of the fat tree and at least one port for a link to a peer server in the indirect hypercube network. Servers are grouped by edge layer network device to form virtual switches in the indirect hypercube network and data packets are routed between servers using routes through the virtual switches. Routes leverage properties of the hypercube topology. Participant servers function as destination points and as virtual interfaces for the virtual switches.

Systems and methods of configuring a computer network are provided. The network can include a first stage of M nodes and a second stage of N nodes. Each node in the first stage can form F communication links with nodes in the second stage and each node in the second stage can form S communication links with nodes in the first stage. Nodes in the first stage are connected to each other only through communication links connecting nodes in the first stage to nodes in the second stage. Communication links between the nodes can be assigned such that the minimum value of the throughput between any pair of nodes in the first stage is maximized.

Achieving balanced interconnectivity in a computer network

The present disclosure describes systems and methods for reconfiguring the links made by a plurality of optical circuit switches between the nodes of the first layer and the nodes of the second layer that reduces the throughput loss when the network transitions form a first logical topology to a second logical topology. More particularly, the first logical topology is realized by a specific physical topology, while the second logical topology may be realized by one or more physical topologies. The disclosure describes a method for selecting a second physical topology from the one or more physical topologies that will realize the second logical topology while reducing the number of links within each of the optical circuit switches that must be reconfigured (i.e, disconnected from their present ports and reconnected to new ports within the optical circuit switch) to transition from the first to second logical topology.

Network fabric reconfiguration

A method, apparatus, and queuing engine implement congestion management. The method may include receiving, via a first interface of the apparatus, data traffic for forwarding to a node of a network. The method may also include receiving, at a second interface of the apparatus, a notification that indicates that congestion is affecting communication with the node, and responsive to the notification, accumulating the data traffic into the queue for a given time period. The method may further include dequeuing the data traffic from the queue after the given time period; and sending the portion of the data traffic to the node via the second interface.

Growing community of inventors

Portola Valley, CA, United States of America

Robert Edman Felderman