Amazon Technologies, Inc.

Michael Brooke Furr

Kevin Christopher Miller

Andrew J Doane

Upendra Bhalchandra Shevade

Mahmoud Ali Abuelela

Diwakar Gupta

Nishant Kanaiyalal Mehta

Kevin P Smith

Aparna Nagargadde

Bashuman Deb

Christopher Ian Hendrie

Joel C Hegg

Ryan David Murphy

Anthony Nicholas Liguori

David Brian Lennon

Po-Chun Chen

Eric Wayne Schultze

Peter N De Santis

Kyle Tailor Akers

Gregory Rustin Rogers

Jeremy T Hall

Sandeep Shantharaj

Leonid Valentinovich Nikolayev

Anish Sukumaran

Nishant Mehta

Methods and apparatus for interfaces to manage direct network peerings. A system may include a data center, endpoint routers and a connectivity coordinator. The coordinator implements a programmatic interface defining connectivity operations. The coordinator receives a request for dedicated connectivity to data center resources, formatted according to the interface. The coordinator selects a target endpoint router at which to establish a physical link to implement the dedicated connectivity, and transmits a response identifying the target endpoint router and including configuration instructions for setting up a physical link for the dedicated connectivity.

Interfaces to manage direct network peerings

A first virtualization control mode to be used for configuring one or more compute instances is determined. Based at least partly on the mode, preparatory configuration operations for launching a compute instance are initiated. The operations include transferring at least a portion of a machine image to a storage device accessible from a host, and initialization of a virtual network interface. A compute instance is instantiated using the machine image at the host. The compute instance is a thread of an intermediary process launched by an administrative agent at the host in response to a launch request.

Computing service with configurable virtualization control levels and accelerated launches

An administrative agent running at a virtualization host of a network-accessible virtualized computing service determines that a first virtual machine is to be instantiated. The agent initiates at least a first configuration operation to enable connectivity for at least a portion of network traffic associated with the first virtual machine. The first configuration operation is performed at least in part using a first virtualization offloading card of the virtualization host. The agent causes a virtualization intermediary process of the virtualization host to launch one or more execution threads of the virtualization intermediary process to implement the first virtual machine. The intermediary process may be swapped to persistent storage, e.g., based on an analysis of resources of the virtualization host.

Network-accessible computing service for micro virtual machines

A number of cells of a control plane of a virtualized computing service are set up, including a first cell with one or more request processing nodes, a local instance of a data store, and metadata indicating a set of virtualization hosts. A request processer transmits a request for a virtual machine to the first cell. A request processor of the cell initiates a workflow to launch the virtual machine using a virtualization host; the workflow includes storing a record of the request in the local instance of the data store.

Autonomous cell-based control plane for scalable virtualized computing

Techniques are disclosed for auditing an IP address prefix that has been assigned to an entity as part of an administrator policy, to determine whether the assignment was implemented on the network. In an embodiment, associations between IP addresses and their assignment are stored in a database. IP addresses are read and semi-authoritative sources (e.g., DNS servers) are queried for information about the IP addresses. Information received in response to the query may be used to validate the IP address (e.g., in a network, all IP addresses used for VM instances will have a corresponding URL in a specific format).

Auditing IP address assignments

A traffic analyzer of a provider network identifies endpoint categories into which traffic directed to or from a first isolated virtual network of the provider network is to be classified. A first endpoint category includes an endpoint configured in a second isolated virtual network. Using packet-level metrics collected at virtualization management components of virtualization hosts, the traffic analyzer determines the amount of data transmitted between the first isolated virtual network and the various endpoint categories during selected time intervals. The traffic analyzer provides the categorized traffic amounts as input to a predictive model, and stores expected future traffic trends generated by the model.

Traffic analyzer for isolated virtual networks

A request to establish a VPN connection between a customer data center and a set of resources of a provider network is received. A new isolated virtual network (IVN) is established to implement a virtual private gateway to be used for the connection. One or more protocol processing engines (PPEs) are instantiated within the IVN, and a respective VPN tunnel is configured between each of the PPEs and the customer data center. Routing information pertaining to the set of resources is provided to the customer data center via at least one of the VPN tunnels, enabling routing of customer data to the set of resources within the provider network from the customer data center.

Virtual private gateways using compute instances

Methods and apparatus for bandwidth metering in large-scale networks are disclosed. Metadata for a network transmission involving a virtualized resource at a host of a provider network, including endpoint address information and a traffic metric, is determined at a metering component. The metadata is aggregated at another metering component and provided to a traffic classification node. The traffic classification node generates a categorized usage record for the network transmission, based at least in part on network topology information associated with the provider network. The categorized usage record is used to determine a billing amount for the network transmission.

Bandwidth metering in large-scale networks

Methods and apparatus for differential bandwidth metering in a network implementing direct peerings. A system includes a plurality of resource collections and a billing manager. The billing manager obtains a first metric of network traffic transmitted on behalf of a client to obtain one or more services from one or more resource collections. The billing manager also obtains a second metric of network traffic transmitted on behalf of the client over one or more private links, where each private link is set up to establish a network path between a respective client network of the client and a respective resource collection of the plurality of resource collections. The billing manager provides composite billing information to the client comprising a differential billing amount dependent upon a difference between the first metric and the second metric.

Differential bandwidth metering for networks with direct peerings

Distributed computer systems prepare and instantiate snapshots by processing sets of nodes representing computer resources that are portions of a larger system. A user may choose to prepare or instantiate a snapshot containing all or only a portion of the data within the distributed computer system, such as a template of the distributed computer system. A template snapshot may be available in an electronic marketplace for other users to purchase and instantiate into a replica of the distributed computer system.

Distributed computer system snapshot instantiation with variable depth

Methods and apparatus for dynamic bandwidth management using routing signals in a network implementing direct peerings. A system includes a resource collection, an endpoint device connected to a private link of a network path to a client network, and a bandwidth manager. The bandwidth manager receives configuration settings of the client, comprising a baseline transmission rate of traffic. The bandwidth manager determines whether a difference between a measured transmission rate of traffic over one or more network paths linking the resource collection and the client network, and a peak transmission capacity of the one or more network paths, exceeds a threshold value. In response to a determination that the difference exceeds the threshold value, the bandwidth manager initiates a transmission of a routing signal to the client network indicative of an available transmission rate that differs from the baseline transmission rate.

Growing community of inventors

Washington, DC, United States of America

Michael Brooke Furr