Amazon Technologies, Inc.

Mustafa Ugur Torun

Mihir Rajendrabhai Patel

Jesse Marcus Dougherty

Timothy William Bray

Marvin Michael Theimer

Benjamin Warren Mercier

Surya Prakash Dhoolam

James Paul Kardos

Bradley Nathaniel Burkett

Hans-Philipp Anton Hussels

Jonathan Patrick Lim

Alireza Sahraei

Curt Richards Carpenter

Kaili Xu

Xinghuang Xu

Arpit Panwar

Yi Fang

Sing Yoong Khew

Soydan Bilgin

Christopher Lee Kubera Harasemchuk

Methods, systems, and computer-readable media for implementing rule-based triggering in a provider network are disclosed. When conditions are met for one or more triggers, one or more rules that comprise the one or more triggers are determined. The conditions are determined in a provider network comprising a plurality of resources. One or more actions are determined in the one or more rules that comprise the one or more triggers. The one or more actions are performed, comprising modifying one or more of the resources in the provider network.

Rule-based action triggering in a provider network

A computing resource monitoring service obtains a request from a third-party partner system to generate a first bus. In response to the request, the computing resource monitoring service updates an account associated with a customer of a remote service to indicate presence of the first bus. The computing resource monitoring service activates the first bus to establish an association with a second bus, obtain a set of events generated by a resource of the third-party partner system, and provides the set of events to the first bus. The computing resource monitoring service publishes the set of events to cause the second bus to distribute the set of events to another remote service.

Hybrid system event processing

Techniques are described for measuring test coverage during phased deployments of software application updates. A phased deployment of a software application update generally involves initially deploying the software update to a subset of compute instances running the software application within a computing environment, using a testing application to verify whether the updated software application is operating as expected, and then deploying the update to the remaining compute instance of the computing environment once the tests are passed. A testing system and framework described herein enables a code deployment service to determine when complete or sufficient test coverage is achieved for software updates deployed in a computing environment supporting a distributed software application.

Measuring test coverage during phased deployments of software updates

Methods, systems, and computer-readable media for generating message envelopes for heterogeneous events are disclosed. A plurality of events are received. The events are associated with computing resources in a multi-tenant environment. The events comprise an event formatted using a content schema and a different event formatted using a different content schema. A message is generated comprising a message envelope and the event. The message envelope comprises a plurality of field names and one or more field values formatted using an envelope schema. A different message is generated comprising a different message envelope and the different event. The different message envelope comprises the field names and one or more different field values formatted using the envelope schema. The message is filtered based at least in part on the message envelope, and the different message is filtered based at least in part on the different message envelope.

Generating message envelopes for heterogeneous events

A software development kit (SDK) is disclosed that can reduce unnecessarily polling of a network service. The SDK is modified to cache response objects received from a network service that describe the state of a computing resource. Polling requests received for the state of the computing resource are satisfied from the cache when possible, thereby avoiding the polling of the network service. Entries in the cache can be invalidated based upon events generated by the network service that indicate that the state of the computing resource has changed.

SDK for reducing unnecessary polling of a network service

Methods, systems, and computer-readable media for implementing a routing handler for rule-based action triggering are disclosed. A request is received at a routing handler from a client. The request specifies a particular operation at a particular one of a plurality of sub-services. The sub-services comprise one or more operations for defining a plurality of rules; the rules are defined to bind triggers to actions. The sub-services also comprise one or more operations for performing the actions in a provider network comprising a plurality of resources. The particular operation in the request is determined at the routing handler. The particular one of the sub-services is determined at the routing handler based on the particular operation, using a routing map. The routing map is dynamically updated when operations provided by sub-services are modified. The request is routed to the particular one of the sub-services.

Routing handler for rule-based action triggering

Devices and processes may perform best-effort key affinity workload sharding. Requests including keys are received at a key based workload manager that applies a consistent sharding algorithm to the keys to select a shard to process the request. For selected shards that are associated with a heavy workload various techniques are performed to iterate though a selection of shards until another shard without a heavy workload is selected. For example, techniques are provided such that for a given request a different shard is selected for each iteration, and shards are selected in the same order for repeated requests with the same key. For keys associated with heavy workloads, shards may be reserved solely for that key. If the techniques fail to find a shard without a heavy workload, a fallback provision may select the shard with the least amount of workload, even if that workload is considered heavy.

Best-effort key affinity workload sharding

A fast and efficient technique to query which records within a table include a list that includes a specified element, where the queried database does not directly support queries on list elements. To support an update, the technique maps individual elements in a list with unique operand values, such as prime numbers, and calculates a function value, such as multiplication, for the list using the unique operand values, where the function value is then associated with the list. To support a query, the technique uses a complementary function, such as a modulo operation, applied to a unique operand for a specified element and to a function value associated with a list to determine if the function value for the list was generated using the unique operand for the specified element.

Growing community of inventors

Seattle, WA, United States of America

Mustafa Ugur Torun