Amazon Technologies, Inc.

Oracle International Corporation

Jacob Adam Gabrielson

Bradley E Marshall

Piragash Velummylum

Joshua Marc Burgin

Michael David Marr

Samuel James McKelvie

Jagwinder Singh Brar

Matthew David Klein

Peter Nicholas Desantis

Ian Paul Nowland

Richendra Khanna

Edward L Wallace

Justin Mason

Allan Henry Vermeulen

Kai Fan Tang

Marcin Piotr Kowalski

Malcolm Featonby

Jonathan B Corley

Nathan Dye

Matthew Paul Baranowski

Zachary J Hansen

Brian L Barker

Brad Bonnett

Eric Jason Brandwine

Diwakar Gupta

Marvin Michael Theimer

James Alfred Gordon Greenfield

Christopher Richard Jacques De Kadt

Nicholas J Maniscalco

Joby Sky Lafky

Roy Mehdi Zeighami

Manish Singh Rathaur

Sumeet Kochar

Vijayaraghavan Apparsundaram

Diane N Lye

Joseph Elmar Magerramov

John Merrill Phillips

Derek Solomon Pai

Leslie Johann Lamprecht

Suresh Kumar

Matthew T Corddry

Andrew James Lusk

Letian Feng

Michael Groenewald

Roberto Pentz De Faria

Robin Alan Golden

Jean-Paul Bauer

Michael Phillip Quinn

Cristian Gabriel Gafton

Marc Andrew Bowes

Michael Benjamin Deardeuff

Wyatt David Camp

Weizhong Hua

Anupama Sharma

Alec Peterson

Elton Victor Pinto

Christopher Sam Goosen

Jeffrey N Peterson

Jacobus Johannes Van Der Merwe

Deepthi Chelupati

Dirk Gerhardes Oberholster

Chi Hai Ho

Christopher Richard Jaques De Kadt

Abhijeet Pradeep Tirthgirikar

Casey Thomas Huggins

Matthew J Eddey

Matthew Paul Baranowki

Terence Lewis

Robert Martin Wise

Christopher Richard De Kadt

Disclosed techniques relate to managing power within a power distribution system. Power consumption corresponding to devices (e.g., servers) that receive power from an upstream device (e.g., a bus bar) may be monitored (e.g., by a service) to determine when power consumption corresponding to those devices has breached or will likely breach a budgeted threshold corresponding to an amount of power allocated to the upstream device. If the budgeted threshold is breached, or is likely to be breached, the service may initiate operations to distribute power caps for the devices and to initiate a timer. Although distributed, the power caps may be ignored by the devices until they are instructed to enforce the power caps (e.g., upon expiration of the timer). This allows the power consumption of the devices to exceed the budgeted power assigned to the upstream device at least until expiration of the timer while avoiding power outage events.

Datacenter level power management with reactive power capping

A real-time data validation system prevents problematic updates from being replicated from a primary database to a replica database. The data validation system may inject a time delay into the replication process, so that when an update in a replication stream fails a specified data validation, the data validation system will prevent the update from being applied to the replica. The replication process may be halted to take snapshot of the replica database during the halt, and resumed after the snapshot. Accordingly, if the problematic update is later confirmed to be invalid, the replica database can be rolled back to a point before the problematic update. Advantageously, the disclosed system can be used to implement a variety of data validations on top of a database replication process to protect the replica database from receiving undesired updates.

Real-time data validation using lagging replica databases

Generally described, systems and methods are provided for monitoring and detecting causes of failures of network paths. The system collects performance information from a plurality of nodes and links in a network, aggregates the collected performance information across paths in the network, processes the aggregated performance information for detecting failures on the paths, analyzes each of the detected failures to determine at least one root cause, and initiates a remedial workflow for the at least one root cause determined. In some aspects, processing the aggregated information may include performing a statistical regression analysis or otherwise solving a set of equations for the performance indications on each of a plurality of paths. In another aspect, the system may also include an interface which makes available for display one or more of the network topology, the collected and aggregated performance information, and indications of the detected failures in the topology.

Monitoring and detecting causes of failures of network paths

Disclosed techniques relate to managing power within a power distribution system. Power consumption corresponding to devices (e.g., servers) that receive power from an upstream device (e.g., a bus bar) may be monitored (e.g., by a service) to determine when power consumption corresponding to those devices breaches (or approaches) a budget threshold corresponding to an amount of power allocated to the upstream device. If the budget threshold is breached, or is likely to be breached, the service may initiate operations to distribute power caps for the devices and to initiate a timer. Although distributed, the power caps may be ignored by the devices until they are instructed to enforce the power caps (e.g., upon expiration of the timer). This allows the power consumption of the devices to exceed the budgeted power associated with the upstream device at least until expiration of the timer while avoiding power outage events.

At a logging service, a representation of a transaction requested by a client is received. The transaction includes at least a first write operation directed to a data object at a first data store. Based at least in part on the contents of the representation and contents of a persistent log of the logging service, a determination is made that the requested transaction is to be committed. A log record corresponding to the first write operation is inserted into the persistent log. The first write is propagated to the first data store after the record has been inserted in the persistent log.

Multi-database log with multi-item transaction support

Techniques are described for enabling users of a cloud provider network to request the conditional termination of computing resources based on a forecasted availability of replacement capacity at a specified time or range of time in the future. A cloud provider network provides an application programming interface that can be used to make such requests, where the computing resources are hosted by the cloud provider network as part of a capacity pool shared by tenants of the cloud provider network. This type of request can be generated, for example, by a user desiring to terminate the use of some number of unproductive computing resources only if the user can be reasonably assured by the cloud provider network that capacity will be available at a future time when the user will likely need the capacity again.

Enabling conditional computing resource terminations based on forecasted capacity availability

Techniques are described for providing service availability zones for computing-related services provided by a cloud provider network. According to some embodiments, the use of service availability zones enables users to achieve greater availability assurance for workloads sensitive to downtime or outages. The sensitivity of such workloads to downtime may be due to financial, regulatory, safety, or other reasons. Such availability-sensitive applications may in some cases be configured to run in multiple regions of a cloud provider network to increase their availability compared to running in a single availability zone. However, for some applications, the implementation of an application across multiple regions may not be feasible or desirable. The use of service availability zones provides user-visible fault domains within a single region of a cloud provider network, thereby enabling users to readily and reliably achieve application layer availability of approximately 99.999% for a service within a single region.

Service availability zones for high availability workloads

A connectivity enablement device includes one or more processors, one or more memories and a hardware input port. The memories store program instructions that when executed examine a token obtained from a token transfer device inserted into the port, and cause one or more messages to be transmitted to a virtualized computing service. The messages indicate (a) the connectivity enablement device, (b) the token transfer device, (c) the token's source and (d) a server. An indication that the server has been configured within an isolated virtual network is obtained at the connectivity enablement device.

Security mechanisms for data plane extensions of provider network services

Disclosed are various embodiments for a system that hardware locks a virtual machine to a host machine. The host machine can include a processor and a memory. A hypervisor can be stored in the memory that, when executed by the processor, causes the computing device to host one or more virtual machines. In addition, machine readable instructions stored in the memory that, when executed by the processor, cause the computing device to at least: receive a request to perform an operation on a virtual machine hosted by the hypervisor; evaluate the request to determine that the request complies with a policy; and in response to a determination that the request complies with the policy, cause the hypervisor to initiate the operation.

Locked virtual machines for high availability workloads

Techniques are described for providing a multi-service storage layer in a cloud provider network for applications and workloads that are highly sensitive to outages affecting 'mission critical' data or other resources. A multi-service storage layer is designed to provide additional resiliency against various types of correlated failures among existing geographic regions by enabling the storage of data using a plurality of separate storage services and storage resource types and across a plurality of regions of the cloud-provider network. A multi-service storage layer provides an application programming interface (API) with actions for storing, retrieving, and querying data stored in a highly available storage resource across a selection of underlying storage services. The ability to store critical data and other resources using a multi-service storage layer provides additional resiliency while also providing a single interface that applications can use to interact with the data stored across any number of underlying storage services.

Multi-service storage layer for storing application-critical data

Techniques are described that enable a cloud provider network to provide specialized regions that can be used to achieve greater availability assurance for workloads highly sensitive to downtime or outages. Cloud provider network users may use specialized regions to complement the use of provider network services offered in other geographic regions defined by the cloud provider network, either to host redundant computing resources or for failover purposes, where the operation of a specialized region is designed to provide additional resiliency against various types of correlated failures among the geographic regions. As one example, a cloud provider network may stage deployments of software updates to the web services provided by the cloud provider network in a manner that ensures that specialized regions receive such updates last and over a relatively long period of time, thereby helping to ensure that any software defects are detected in an earlier deployment of the update.

Specialized cloud provider regions for availability-sensitive workloads

Systems and methods are disclosed to implement a real-time data validation system that prevents problematic updates from being replicated from a primary database to a replica database. In embodiments, the data validation system injects a time delay into the replication process, so that when an update in a replication stream fails a specified data validation, the data validation system will prevent the update from being applied to the replica. In embodiments, the replication process is halted, a snapshot of the replica database is taken during the halt, and replication to the replica is resumed. Accordingly, if the problematic update is later confirmed to be invalid, the replica database can be rolled back to a point before the problematic update. Advantageously, the disclosed system can be used to implement a variety of data validations on top of a database replication process to protect the replica database from receiving undesired updates.

Techniques are described for optimizing the allocation of computing resources provided by a service provider network&#x2014;for example, compute resources such as virtual machine (VM) instances, containers, standalone servers, and possibly other types of computing resources&#x2014;among computing workloads associated with a user or group of users of the service provider network. A service provider network provides various tools and interfaces to help businesses and other organizations optimize the utilization of computing resource pools obtained by the organizations from the service provider network, including the ability to efficiently schedule use of the resources among workloads having varying resource demands, usage patterns, relative priorities, execution deadlines, or combinations thereof. A service provider network further provides various graphical user interfaces (GUIs) to help users visualize and manage the historical and scheduled uses of computing resources by users' workloads according to user preferences.

Coordinated predictive autoscaling of virtualized resource groups

Growing community of inventors

Seattle, WA, United States of America

Jacob Adam Gabrielson