Equinix, Inc.

Danjue Li

Ankur Sharma

Lanfa Wang

Yury Kamen

Mustafa Arisoylu

Yakov Kamen

Muhammad Durrani

Imam Sheikh

Chen Xi

Kaladhar Voruganti

Brian James Lillie

Syed Hashim Iqbal

Rizwan Jamal

Anand Ozarkar

Alex Wilms

Ravi Kiran Pasula

Rodney Martin Elder

Sreekanth Narayanan

Lance Weaver

David Gofman

Stanley Chernavsky

Edward Rosinzonsky

Zoe Liu

Christopher Alan Anderson

In general, this disclosure describes techniques for configuring and provisioning, with a distributed artificial intelligence (AI) fabric controller, network resources in an AI fabric for use by AI applications. In one example, the AI fabric controller is configured to discover available resources communicatively coupled to a cloud exchange; obtain a set of candidate solutions, each candidate solution of the set of candidate solutions comprising an AI application and a configuration of resources for use by the AI application; filter, based on one or more execution metrics corresponding to each of the candidate solutions, the set of candidate solutions to generate a filtered set of candidate solutions; generate provisioning scripts for the filtered set of candidate solutions; execute the provisioning scripts to provision resources for each candidate solution in the filtered set of candidate solutions; and create an execution environment for each candidate solution in the filtered set of candidate solutions.

Distributed artificial intelligence fabric controller

Techniques are disclosed for performing time synchronization at a plurality of computing devices in a network. In one example, a method comprising obtaining timestamp data in accordance with a synchronization operation for a timing protocol; computing a skewness estimate and an offset estimate from the timestamp data by executing a regression analysis, wherein the regression analysis is configured to train a first model to predict the offset estimate and the skewness estimate, the offset estimate comprising a clock time difference between the first clock and the second clock; computing a corrected skewness estimate and a corrected offset estimate based on a second model having parameters based on the offset estimate and the skewness estimate; and modifying a current time value of at least one of the first clock or the second clock based on at least one of the corrected offset estimate or the corrected skewness estimate.

Time synchronization using model correction

This disclosure describes techniques that include interpreting information about a smart contract so that the terms of the smart contract can be implemented across a diverse array of consensus networks or blockchain platforms. In one example, this disclosure describes a method that includes receiving, by a first computing device, information describing a smart contract, wherein the first computing device is included within a first plurality of computing devices, each on a first consensus network that maintains a first distributed ledger; performing, by the first computing device and in response to receiving the information describing the smart contract, operations to update the first distributed ledger; and interpreting, by at least one of the first plurality of computing devices, the information describing the smart contract to determine and perform at least one of a plurality of first smart contract operations on the first consensus network.

Smart contract interpreter

In general, techniques are described for a hierarchical, distributed DHCP system for managing IP address assignment among distributed networks of computing devices. For example, a system may include a central DHCP server configured to manage a plurality of distributed DHCP servers, each distributed DHCP server configured to perform DHCP using IP addresses allocated from a common prefix for a tenant associated with computing devices managed by multiple DHCP servers. The central DHCP server allocates IP addresses to the distributed DHCP servers, e.g., on an on-demand basis from the common pool and may handle concurrent requests for IP addresses from distributed DHCP servers. Each of the distributed DHCP servers may store records for IP addresses and media access control (MAC) addresses for computing devices managed by that distributed DHCP server, and the DHCP servers may send these records to the central DHCP server to facilitate IP assignment coherency.

Hierarchical distributed dynamic host configuration protocol system

Techniques are disclosed for performing time synchronization for a plurality of computing devices without relying upon a minimum measured delay. In one example, processing circuitry obtains time stamp data in accordance with an iteration of a synchronization operation for a timing protocol, wherein the time stamp data describes one or more measured delays for a path between a first computing device and a second computing device, computes a skewness estimate from the time stamp data using a regression analysis, the skewness estimate comprising a frequency difference between a first clock at the first computing device and a second clock at the second computing device, computes an offset estimate between the first clock and the second clock by applying a prediction model to the skewness estimate; and corrects at least one of the first clock or the second clock based at least on the offset estimate.

Time synchronization using skew estimation

The disclosure describes methods and systems for performing time synchronization in a heterogeneous system. In one example, a method includes, for each secondary device of one or more secondary devices in a network, determining, by a computing system, one or more time synchronization characteristics for the respective secondary device; and generating, by the computing system and based on at least the respective one or more time synchronization characteristics for each respective secondary device of the one or more secondary devices in the network, a time synchronization report for the network, wherein the one or more time synchronization characteristics include health data for the one or more secondary device.

Clock synchronization in a heterogeneous system

The disclosure describes methods and systems for performing time synchronization in a heterogeneous system. In one example, a method includes evaluating, by a computing system, one or more network conditions of a network to determine whether to perform a time synchronization process with a secondary device in the network, wherein the one or more network conditions include a health score for the secondary device, and, in response to determining, based on the evaluation of the one or more network conditions, to perform the time synchronization process: determining based at least in part on a time indication for a clock on a master device and a time indication for a clock on the secondary device, a time synchronization offset for the secondary device; and sending the time synchronization offset for the secondary device to the secondary device in a data packet.

This disclosure describes techniques that include use of a distributed ledger to arrange for and/or initiate provisioning of network services, and also to validate payment for such network services. In one example, this disclosure describes a method that includes modifying, by a computing system, a distributed ledger maintained by a consensus network to implement a smart contract that is configured to generate, in response to receiving a request for network services, a provisioning request; receiving the provisioning request, by the computing system and from a computing device on the consensus network that is executing the smart contract, wherein the provisioning request includes information describing requested network services; and provisioning, by the computing system and based on the information describing the requested network services, network services.

Efficient network service provisioning

Techniques are disclosed for performing time synchronization for a plurality of computing devices that exhibit asymmetric path delay. In one example, processing circuitry receives data indicative of a graph comprising a plurality of nodes and vertices, wherein each node represents a clock and each vertex represents a bidirectional path between two clocks. Each bidirectional path has a first path delay in a first direction that is different from a second path delay in a second direction. The processing circuitry determines one or more closed loops in the graph and a path delay of the closed loop. The processing circuitry applies a minimization function to the path delay of each closed loop to determine values for the first and second path delays of each bidirectional path. The processing circuitry applies, based on the values for the first and second path delays of each bidirectional path, a time correction to a clock.

Timing synchronization for clock systems with asymmetric path delay

The disclosure describes methods and systems for performing time synchronization in a heterogeneous system. In one example, a method includes, for each secondary device of a plurality of secondary devices in a network: determining, by a computing system and based at least in part on a time indication for a clock on a master device and a time indication for a clock on a secondary device in the network, a time synchronization offset for the respective secondary device; collecting, from the respective secondary device, one or more static parameters and one or more dynamic parameters; determining, based on the one or more parameters, a weight to associate with the time synchronization offset for the respective secondary device; determining, based on each of the respective time synchronization offset for each of the plurality of secondary devices and the respective associated weight, a universal time synchronization offset for the network.

This disclosure describes techniques for delivering high-accuracy and high-precision clock synchronization in heterogeneous distributed computer clusters. For example, the disclosure describes a synchronization engine that sets efficient clock synchronization processes based on a cluster node's characteristics, pricing, precision, geolocation, and/or cluster topology, while in some cases using a combination of master clock data with internal atomic clocks of computers. The techniques described herein integrate the synchronization engine into a time synchronization process that may provide stability, versatility, precision and cost balance using technical improvements for characterizing timing system delivery channels.

Timing synchronization service and distribution system

In one example, a method includes receiving, by a first network device via a routing protocol peering session with a peer router in a first autonomous system, a plurality of routing protocol routes to destination addresses, each routing protocol route specifying a network address prefix and an identifier of the autonomous system that originated the routing protocol route; receiving network address prefix ownership information from a distributed ledger storing a plurality of associations between respective network address prefixes and respective autonomous system identifiers of autonomous systems confirmed to own the respective network address prefixes; determining, based at least on the prefix ownership information, whether any of the plurality of routing protocol routes specifies an autonomous system identifier different than specified by the associations; and in response to determining that one of the routes specifies an autonomous system identifier different than specified by the plurality of associations, performing an action.

Growing community of inventors

Dublin, CA, United States of America

Danjue Li