Amazon Technologies, Inc.

Jacob Carr

Brian Thomas Kachmarck

Gracjan Polak

Kanika Kalra

Amit Chhabra

Chaiwat Shuetrakoonpaiboon

Chen Zhou

Jazarine Jamal

Muhammad Usman

Stanislav Pavlovskii

Syed Sajid Nizami

Ryan Preston Gantt

Eric Jason Brandwine

Gaurav Gupta

Christopher Richard Jacques De Kadt

Aaron Ben Fernandes

Morgan Akers

James Hsiaosung Chuang

Naveen Anand Subramaniam

Asad Khan Durrani

Vinayak Sundar Raghuvamshi

Per Weinberger

Nimit Kumar Garg

Reza Shahidi-Nejad

Brandon William Porter

Zachary Ganwise Fewtrell

Ethan Fitzgerald John

Stuart Olmstead-Wilcox

Evan Corriere

Sean Robert Connell

Jesse Morgan

Mohsen Azimi

Andrew K Samnick

Swapneil Deshmukh

Andrew Samnick

Abhijit Prakash Bhatnagar

Yusof Ganji

Jason Adonis Timmons

Xinrui Li

Sahil Sharma

Tristan Niles Cecil

Huaqing Fang

Gregory Michael Li

Techniques are disclosed to implement an adaptive testing service (ATS) capable of automatically generating test cases for a network service to adapt test coverage to observed behaviors of the network service. In embodiments, the ATS uses telemetry data from a production version of the network service to identify classes of testable behaviors. Test cases are generated for the behaviors and assigned weights based on frequency or recency metrics of the behaviors. The test cases are stored in a test case repository, and may be used to monitor the production version of the network service or verify code changes to a development version of the network service. The test case weights may be used to select which test cases to run or determine whether code changes should be accepted or rejected. The test cases are evolved over time to adapt to behavior changes in the network service.

Adaptive testing service that generates test cases from observed behaviors

Layer-specific modifications may be applied to a network communication received at a proxy according to configurable rules at the proxy. A modification configuration may be received at the proxy that corresponds to a layer of a communication stack supported by the proxy. When a communication is received, data from the layer of the communication stack of the network communication may be evaluated to determine whether a modification is applicable to the communication according to the modification configuration. An applicable modification may be applied by proxy as part of processing the communication at the proxy.

Layer-specific modifications to network communications according to configurable rules at a proxy

Technologies are described for cost based management of cache entries stored in a computer memory. In one example, a plurality of cache entries may be stored at a cache in a computer memory and the cache entries may have a cost measure associated with individual cache entries. A cost measure may represent a computing cost of an application to generate a cache entry. An incoming cache entry may be received at the cache, where the incoming cache entry has a cost measure associated with the incoming cache entry. In response to receiving the incoming cache entry, a cache entry that has a lower cost measure than the cost measure for other cache entries may be identified for eviction from the cache. The cache entry identified for eviction may be evicted from the cache, and the incoming cache entry may be written into the cache stored in the computer memory.

Cost based cache eviction

Methods, systems, and computer-readable media for a cell-based storage system with failure isolation are disclosed. A first subset of storage nodes is selected from a set of storage nodes of a data store and assigned to a first partition of data. Replicas of the first partition are stored using the first subset. A second subset of storage nodes is selected from the set of storage nodes and assigned to a second partition of data. The second subset is selected based (at least in part) on the membership of the first subset, and the second subset comprises at least one node not present in the first. Replicas of the second partition are stored using the second subset. Access requests associated with the first partition are routed to the first subset of storage nodes. Access requests associated with the second partition are routed to the second subset of storage nodes.

Cell-based storage system with failure isolation

Methods, systems, and computer-readable media for augmenting storage functionality using emulation of storage characteristics are disclosed. An access request for a data set is received. The access request is formatted according to a first protocol associated with a first data store, and the first data store is associated with first storage characteristics. The access request is translated into a translated access request. The translated access request is formatted according to a second protocol associated with a second data store, and the second data store is associated with second storage characteristics differing at least in part from the first storage characteristics. The translated access request is sent to the second data store. The translated access request is performed by the second data store on the data set using emulation of one or more of the first storage characteristics not included in the second storage characteristics.

Augmenting storage functionality using emulation of storage characteristics

Modifications to caching performed between different services may be determined. A history of requests and responses between the different services may be obtained. The history may be evaluated to determine respective frequencies of parameters between the services. The frequencies of parameters may be evaluated to determine one or more modifications to caching among the different services. The modifications may be provided in order to be applied to change caching performance for subsequent requests.

Modifying caching amongst services from a history of requests and responses

Methods, systems, and computer-readable media for record-by-record live migration using segmentation are disclosed. Migration of a data set comprises, for a record in a segment being migrated, storing a first status indicating that the record is offline in a source data store. An instance of the record is stored in the destination data store, and a second status is stored to indicate that the record is online in the destination. The record is deleted from the source after the second status is stored. During the migration, a read request for the record is received and determined to be associated with the segment being migrated. A response to the read request is generated that comprises an authoritative instance of the record. The instance of the record in the destination is determined to represent the authoritative instance based (at least in part) on the first status and the second status.

Record-by-record live migration using segmentation

Using emulation of storage characteristics to perform an access request translated between different protocols

Methods, systems, and computer-readable media for segmentation of maintenance on distributed systems are disclosed. A data set is partitioned according to a hash function into a plurality of segments. A maintenance activity is initiated on a first segment. During the maintenance activity, a first request to perform a first action on the data set is received. Based at least in part on determining that the first request is associated with the first segment using the hash function, the first action is performed using additional processing associated with the maintenance activity. During the maintenance activity, a second request to perform a second action on the data set is received. Based at least in part on determining that the second request is associated with a second segment using the hash function, the second action is performed without the additional processing associated with the maintenance activity.

Segmentation of maintenance on distributed systems

Methods, systems, and computer-readable media for transactionally consistent backup of partitioned storage are disclosed. Backup operations are initiated for a first and second partition of one or more data stores. The position of a first marker is determined in a first log associated with the first partition. The first log indicates a sequence of changes for the first partition. The position of a second marker is determined in a second log associated with the second partition. The second log indicates a sequence of changes for the second partition. The first or second marker is determined to violate a transactional boundary across the first and second partitions. A position of at least one of the markers is modified such that the markers do not violate the transactional boundary.

Transactionally consistent backup of partitioned storage

Items within a data set can be redistributed to different partitions of a distributed data store according to modified hash values. A redistribution event may, in some embodiments, trigger the redistribution of a data set. Modified hash values for items in the data set may be generated according to a secondary hashing scheme to be used for identifying the location of a new partition to store the item instead of hash values based on current key values that identify a current partition that stores the item. Access can be provided to items based on the modified hash values to determine the new partition that stores the item after redistribution.

Redistributing a data set amongst partitions according to a secondary hashing scheme

Methods, systems, and computer-readable media for implementing augmenting storage functionality using composition are disclosed. A logical data store is determined based at least in part on storage requirements specified by a user. The logical data store comprises a composition of a plurality of storage adapters for a plurality of physical data stores including a first storage adapter for a first physical data store and a second storage adapter for a second physical data store. The first storage adapter and the second storage adapter are associated with different storage characteristics. A plurality of storage requests for the logical data store are received. Individual ones of the storage requests are routed to the first physical data store using the first storage adapter. Individual ones of the storage requests are routed to the second physical data store using the second storage adapter.

Augmenting storage functionality using composition

A computing resource service provider may provide customers with access to various services provided by a set of service endpoints. During certain intervals of time increased network traffic may cause an increase in the consumption of computing resources of the set of service endpoints. A throttling trigger may be detected that may cause one or more routing devices to reroute network traffic to one or more lower tier routing devices. The lower tier routing devices may maintain throttling information and forward network traffic to the set of service endpoints.

Shaping network traffic using throttling decisions

Aspects of provisioning computing units based on improved failure resiliency are described. In one embodiment, an infrastructure component shared between a pair of computing units is identified. A failure rate for the infrastructure component is obtained, and a failure probability for a class of assigned computing units is computed based in part on the failure rate. A spread request related to the class of assigned computing units is also received. In response to the spread request, an altered composition of computing units is determined, and a difference between a failure probability for the altered composition of computing units and the failure probability for the class of assigned computing units is computed. In one embodiment, when a spread score improvement value associated with the difference meets a spread criteria of the spread request, the altered composition of computing units may be provisioned for use.

Failure resiliency provisioning

Growing community of inventors

Seattle, WA, United States of America

Jacob Carr