公开(公告)号：US09563845B1

公开(公告)日：2017-02-07

申请号：US14297014

申请日：2014-06-05

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： Brandon William Porter ,  Robert Stanley Bailes ,  Aaron Ben Fernandes ,  Michael James McInerny ,  Vishal Parakh ,  Harsha Ramalingam ,  Karthik Tamilmani

IPC: H04L12/26 ,  G06N5/02 ,  H04L29/08 ,  G06F11/34

CPC classification number: G06N5/022 ,  G06F9/00 ,  G06F11/3438 ,  G06F17/30867 ,  H04L67/22

Abstract: Techniques are described for employing precomputed results of applying rules to content items, the rules applicable to determine whether content items may be electronically published. On receiving a request for a content item, rules applicable to the content item may be identified. A datastore of precomputed results of rule application may be accessed to determine whether the datastore includes a result of applying a current or previous version of each rule. If the datastore includes a current result, the current result may be employed in determine whether the content item may be presented. If the datastore includes a previous result, the previous result may be so employed. If the datastore includes a previous result or no appropriate result, a job may be queued to calculate the current result asynchronously relative to the request for the content item.

Abstract translation: 描述了使用将规则应用于内容项的预计算结果的技术，适用于确定内容项目是否可以被电子发布的规则。 在接收对内容项目的请求时，可以识别适用于内容项目的规则。 可以访问规则应用的预计算结果的数据存储区以确定数据存储是否包括应用每个规则的当前版本或先前版本的结果。 如果数据存储区包括当前结果，则可以使用当前结果来确定是否可以呈现内容项目。 如果数据存储包含先前的结果，则可能会使用以前的结果。 如果数据存储包含先前的结果或没有适当的结果，则作业可能被排队，以相对于内容项的请求异步计算当前结果。

公开(公告)号：US10152239B1

公开(公告)日：2018-12-11

申请号：US15051560

申请日：2016-02-23

Applicant: Amazon Technologies, Inc.

Inventor： Karthik Tamilmani

IPC: G06F3/00 ,  G06F3/06 ,  G06F11/14

Abstract: Methods, systems, and computer-readable media for migration to a two-tiered data store are disclosed. A destination data store comprises a primary (warm) tier and a secondary (cold) tier having higher latency for data access than the primary tier. Migration is initiated of live updates from a source data store to the primary tier of the destination data store. A replica of the source data store is generated while the source data store is online. Transform logic is optionally applied to source data from the replica of the source data store. The source data (transformed or not) is imported to the secondary tier of the destination data store.

公开(公告)号：US09189515B1

公开(公告)日：2015-11-17

申请号：US13791453

申请日：2013-03-08

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： Karthik Tamilmani ,  Santosh Kalki

IPC: G06F17/30

CPC classification number: G06F17/30445 ,  G06F17/30424 ,  G06F17/30943

Abstract: Techniques are described for retrieving data stored in disparate datastores that support different or heterogeneous storage systems. A report description may be received from a user, the report description including multiple query templates for generating queries to retrieve data from the disparate datastores. The report description may be analyzed to determine input parameters for generating the queries. A user interface may be dynamically generated to solicit input values corresponding to the input parameters. On receiving the input values, the system may generate and execute the queries of the query plan, and combine the results based on result combination information included in the report description.

Abstract translation: 描述了用于检索存储在支持不同或异构存储系统的不同数据存储中的数据的技术。 可以从用户接收报告描述，报告描述包括用于生成查询以从不同数据存储中检索数据的多个查询模板。 可以分析报告描述以确定用于生成查询的输入参数。 可以动态地生成用户界面以征求对应于输入参数的输入值。 在接收输入值时，系统可以生成并执行查询计划的查询，并根据报告描述中包含的结果组合信息组合结果。

公开(公告)号：US09230011B1

公开(公告)日：2016-01-05

申请号：US13689919

申请日：2012-11-30

Applicant: Amazon Technologies, Inc.

Inventor： Karthik Tamilmani

IPC: G06F17/30

CPC classification number: G06F17/30613

Abstract: Methods and systems for index-based querying of archived data sets are disclosed. A plurality of indices are generated, each comprising a plurality of pointers to storage locations of a plurality of updates to a document. A query request is received after generating the plurality of indices. The query request comprises an identifier of the document and a specified time range. A subset of the indices that comprise the identifier of the document within the specified time range are determined. From the subset of the indices are retrieved a subset of the pointers to the storage locations for one or more of the updates to the document within the specified time range. The one or more updates to the document within the specified time range are retrieved using the subset of the plurality of pointers.

Abstract translation: 公开了用于存档数据集的基于索引的查询的方法和系统。 生成多个索引，每个索引包括多个指向文档的多个更新的存储位置的指针。 在生成多个索引之后接收查询请求。 查询请求包括文档的标识符和指定的时间范围。 确定在指定时间范围内构成文档的标识符的索引的子集。 从索引的子集中检索在指定时间范围内对文档的一个或多个更新的存储位置的指针的子集。 使用多个指针的子集来检索指定时间范围内的文档的一个或多个更新。