公开(公告)号：US20170228290A1

公开(公告)日：2017-08-10

申请号：US15495827

申请日：2017-04-24

Applicant: Amazon Technologies, Inc.

Inventor： Maximliano Maccanti ,  Timothy Andrew Rath ,  Rama Krishna Sandeep Pokkunuri ,  Akshat Vig ,  Clarence Wing Yin NG ,  Srivaths Badrinath Copparam ,  Rajaprabhu Thiruchi Loganathan ,  Wei Xiao ,  William Alexander Stevenson

IPC: G06F11/14

CPC classification number: G06F11/1451 ,  G06F11/1458 ,  G06F11/1464 ,  G06F11/1469 ,  G06F11/2094 ,  G06F2201/80

Abstract: A system that implements a data storage service may store data for a database table in multiple replicated partitions on respective storage nodes. In response to a request to back up a table, the service may back up individual partitions of the table to a remote storage system independently and (in some cases) in parallel, and may update (or create) and store metadata about the table and its partitions on storage nodes of the data storage service and/or in the remote storage system. Backing up each partition may include exporting it from the database in which the table is stored, packaging and compressing the exported partition for upload, and uploading the exported, packaged, and compressed partition to the remote storage system. The remote storage system may be a key-value durable storage system in which each backed-up partition is accessible using its partition identifier as the key.

公开(公告)号：US20170154091A1

公开(公告)日：2017-06-01

申请号：US15431706

申请日：2017-02-13

Applicant: Amazon Technologies, Inc.

Inventor： Akshat Vig ,  Timothy Andrew Rath ,  Stuart Henry Seelye Marshall ,  Rande A. Blackman ,  David Alan Lutz ,  Jian Wang ,  Jiandan Zheng ,  Janani Narayanan

IPC: G06F17/30

Abstract: Methods and apparatus for conditional master election in a distributed database are described. A plurality of replicas of a database object are stored by a distributed database service. Some types of operations corresponding to client requests directed at the database object are to be coordinated by a master replica. Client access to the database object is enabled prior to election of a master replica. In response to a triggering condition, a particular replica is elected master. The master coordinates implementation of operations with one or more other replicas in response to client requests.

公开(公告)号：US09578080B1

公开(公告)日：2017-02-21

申请号：US14223952

申请日：2014-03-24

Applicant: Amazon Technologies, Inc.

Inventor： Allan H. Vermeulen ,  Timothy Andrew Rath

IPC: G06F15/16 ,  H04L29/08

CPC classification number: H04L67/02 ,  G06F9/5011 ,  G06F2209/504 ,  H04L47/82 ,  H04L67/1008 ,  H04L67/42 ,  Y02D10/22

Abstract: Resource management techniques for shared resources in a distributed system are described. Clients and servers may exchange messages according to an asynchronous messaging protocol that does not guarantee delivery or ordering of messages. A client may send a resource request message including a client timestamp and a measure of client resource demand. The server may allocate a grant of the resource to the client in a manner that prevents resource overload, and indicate the grant to the client via a message including a logical timestamp, the amount of resource granted, the client's original timestamp, and a grant expiration time. The server may acknowledge the grant and cooperatively use the resource in accordance with the grant's terms.

Abstract translation: 描述了分布式系统中共享资源的资源管理技术。 客户端和服务器可以根据不保证消息的传递或排序的异步消息协议来交换消息。 客户端可以发送包括客户端时间戳和客户端资源需求量度的资源请求消息。 服务器可以以防止资源过载的方式向客户端分配资源的许可，并且通过包括逻辑时间戳，所授予的资源量，客户端的原始时间戳和授权到期的消息来指示给客户机 时间。 服务器可以根据授权条款确认授权并合作使用资源。

公开(公告)号：US09569513B1

公开(公告)日：2017-02-14

申请号：US14023165

申请日：2013-09-10

Applicant: Amazon Technologies, Inc.

Inventor： Akshat Vig ,  Timothy Andrew Rath ,  Stuart Henry Seelye Marshall ,  Rande A. Blackman ,  David Alan Lutz ,  Jian Wang ,  Jiandan Zheng ,  Janani Narayanan

IPC: G06F7/00 ,  G06F17/30

CPC classification number: G06F17/30575 ,  G06F17/3051 ,  G06F17/30545 ,  G06F17/30587 ,  G06F17/30943

Abstract: Methods and apparatus for conditional master election in a distributed database are described. A plurality of replicas of a database object are stored by a distributed database service. Some types of operations corresponding to client requests directed at the database object are to be coordinated by a master replica. Client access to the database object is enabled prior to election of a master replica. In response to a triggering condition, a particular replica is elected master. The master coordinates implementation of operations with one or more other replicas in response to client requests.

Abstract translation: 描述了分布式数据库中条件主选择的方法和设备。 数据库对象的多个副本由分布式数据库服务存储。 对应于针对数据库对象的客户端请求的某些类型的操作将由主副本进行协调。 在选择主副本之前，客户端对数据库对象的访问已启用。 响应于触发条件，特定副本被选为主控。 主人根据客户端请求协调与一个或多个其他副本的操作执行。

公开(公告)号：US08812897B1

公开(公告)日：2014-08-19

申请号：US13921938

申请日：2013-06-19

Applicant: Amazon Technologies, Inc.

Inventor： Michael T. Helmick ,  Jakub Kulesza ,  Timothy Andrew Rath ,  Stefano Stefani ,  David Alan Lutz

IPC: G06F11/00

CPC classification number: G06F11/2025 ,  G06F11/00 ,  G06F11/2028 ,  G06F11/2041 ,  G06F11/2046 ,  G06F11/2097

Abstract: Disclosed are various embodiments for distributing data items within a plurality of nodes. A data item update request is replicated from a master node in a plurality of nodes to a plurality of slave nodes within the plurality of nodes. The replicated data item update request is determined to be locality-based durable. Responsive to the determination that the replicated data item update request is locality-based durable, the data item update request is confirmed to a client, wherein the client had originated the data item update request. Upon failover of the master node to another node within the plurality of nodes, a fault-tolerant failover quorum ensures that all previously confirmed updates are found and recognized by the new master node.

Abstract translation: 公开了用于在多个节点内分发数据项的各种实施例。 将数据项更新请求从多个节点中的主节点复制到多个节点内的多个从节点。 复制的数据项更新请求被确定为基于局部性的持久性。 响应于确定复制数据项更新请求是基于局部性的持久性，数据项更新请求被确认给客户端，其中客户端已经发起数据项更新请求。 在主节点故障转移到多个节点中的另一节点之前，容错故障转移仲裁确保所有先前确认的更新都被新的主节点找到并被识别。