Amazon Technologies, Inc.

Alibaba Group Holding Limited

Yijun Lu

Wei Xiao

Kiran-Kumar Muniswamy-Reddy

Stuart Henry Seelye Marshall

Miguel Mascarenhas Filipe

Bjorn Patrick Swift

Stefano Stefani

James Robert Hamilton

Yunfeng Zhu

Yunfeng Tao

Zhiyang Tang

Timothy Andrew Rath

Xianglong Huang

Jiandan Zheng

Jun Song

Dongbai Guo

Yanchao Li

Jun Yu

Jibin Lei

Jia Feng

David Craig Yanacek

Rande Anthony Blackman

Barry Bailey Hunter, Jr

Lin Cheng

Youquan Feng

Stuart Henry Marshall

Akshat Vig

Ying Chu Wang

Somasundaram Perianayagam

Qiang Liu

Bo Yang

Adam Douglas Morley

Pejus Manoj Das

Kiran Kumar Muniswamy Reddy

Yi Liu

Ling Xu

Hong Tang

Hong Zhou

Lingling Jin

Aanchal Gupta

Zhengxiong Tian

Shirley Xue Yi Ly

Ying Lin

Lin Yu

Guan Wang

Jiangwei Huang

Nan Zhang

Shiwei Li

Xiaolin Meng

Yu Zhou

Chao Qian

Kaige An

Shuiwang Liu

Jun Ruan

Haoran Yang

Ying Wang

Apparatuses and methods for providing resources are provided that include receiving power statuses of resources of a system capable of providing the resources; quantifying the power statuses of the resources; calculating an available soft capacity of the system based on the quantified power statuses and a total capacity of the system; and providing an assigning amount of the resources beyond the calculated available soft capacity to one or more users.

Power management method

System and methods may include receiving a request for assigning resources to a first application associated with a first entity, the first application having a plurality of levels of performance each corresponding to a different power consumption; and determining a particular level of performance of the first application to run on the resource under a first constraint that a sum of a power consumption of the first application running at the particular level of performance and power consumptions of one or more other resources running other applications associated with the first entity is less than or equal to a power budget associated with the first entity.

Unified power management

Systems and methods for a hybrid system-on-chip usable for predicting performance and power requirements of a host server include a big cores module, including central processing units, for receiving and pre-processing performance and power metrics data of the host server and to allocate computing resources, a small cores module, including massively parallel processing units, for mapping each instance associated with host server in the performance and power metrics data to a corresponding massively parallel processing unit based on the allocated computing resources for a per-instance metrics calculation, and an artificial intelligence (AI) accelerator for calculating performance and power prediction results based on the per-instance calculations from the small cores module.

Hybrid system-on-chip for power and performance prediction and control

A distributed data storage system may implement determining indexing progress for a table stored in the distributed data storage system. A table may be stored in multiple table partitions. When a secondary index is created for the table, each partition may independently index the items stored within the table partition in order to identify those items in the partition that should be stored in the secondary index. During creation of the secondary index, creation progress points that indicate the progress of the indexing across the table partitions may be determined. The creation progress points may be provided via a programmatic, textual, or graphical interface to the distributed data storage system.

Determining indexing progress for a table in a distributed data store

One or more table partitions may communicate with an index partition that may be a master of a replication group. A communications channel may exist between table partitions and the index partition. Upon splitting the index partition, communications between the table partitions and the index partition may be suspended. Upon completion of the split, communications may be reestablished between the table partitions and a partition, of the replication group of index partitions, designated to be a master following the split. Messages accumulated by the table partitions during the split may be sent to the index partition upon reestablishing communications.

Table and index communications channels

A method and device for allocating a distributed lock is provided. A globally unique service process ID (SPI) is introduced for each distributed application service process while guaranteeing correctness of a distributed lock, and the SPI is used to directly manage ownership of the distributed lock. The service replacement process sends to the server a request for inheriting ownership of the distributed lock before the lifecycle of the lock file expires, the inherit request including the SPI.

Allocation method and device for a distributed lock

Embodiments of the present disclosure provide a method comprising: adding a second set of one or more servers to a first set of servers of a consensus system to form an expanded set of servers, performing a capacity-expanding operation on one of the second set of one or more servers. The capacity-expanding operation comprises: pausing a service process related to distributed services, updating configuration information of the first set of servers in response to the second set of one or more servers being added, and resuming the service process and initiating master server election for the expanded set of servers, wherein one of the expanded set of servers is elected as a master server, the rest of the expanded set of servers are considered as slave servers. The method can also include performing the capacity-expanding operation sequentially on each of the slave servers, and subsequently, performing the capacity-expanding operation on the master server, and performing the capacity-expanding operation on the second set of one or more servers that have not been subjected to the capacity-expanding operation after completion of the capacity-expanding for the first set of servers.

Online capacity-expanding and online capacity-reducing methods and apparatuses for distributed consensus system

Methods and apparatus for token-based admission control for replicated writes are disclosed. Data objects are divided into partitions, and corresponding to each partition, at least a master replica and a slave replica are stored. A determination as to whether to accept a write request directed to the partition is made based at least in part on one or more of (a) available throughput capacity at the master replica, and (b) an indication, obtained using a token-based protocol, of available throughput capacity at the slave replica. If the write request is accepted, one or more data modification operations are initiated.

Token-based admission control for replicated writes

Providers of web services and other types of software as a service may be subject to service-level agreements requiring that response times be within a defined range. For efficiency, multiple services may be hosted on the same set of computing nodes, which may jeopardize adherence to service-level agreements. A control system may involve classifying service requests and determining desired values for measurements such as latency. An error value may be calculated based on the difference between measured and desired values. A controller may adjust a rate of capacity utilization for the computing nodes based on the current error, a history of past errors, and a prediction of future errors.

Real-time monitoring of IO load and latency

A pricing policy to be applied to token population changes at a token bucket used for admission control during burst-mode operations at a work target is determined. Over a time period, changes to the token population of that bucket are recorded. An amount to be charged to a client is determined, based on the recorded changes in token population and an associated pricing amount indicated in the policy.

Token-based policies burst-mode operations

A lease-based heartbeat protocol method is provided. The method may include sending a heartbeat request to a server device in a lease period, and receiving a heartbeat request response from the server device; and determining a retry sending at adaptive interval in response to the heartbeat request response being abnormal, and sending a retry heartbeat request to the server device again after the retry sending interval is past, until the lease period expires or a corresponding heartbeat request response is normal. As such, two successive retry heartbeat requests can be sent at a relatively large time interval at an initial stage of heartbeat request retry. At a later stage of the heartbeat request retry, the time interval associated with the retry heartbeat requests is reduced, such that re-sent heartbeat requests can be sent at a higher speed.

Lease-based heartbeat protocol method and apparatus

A system, apparatus, and method for maintaining the consistency of global information is disclosed herein. In one embodiment, the method includes retrieving current version information associated with the global information from a global information server; retrieving global information from the global information server based on the current version information; updating a period of validity of the current version information based on a status of communication with the global information server; and suspending a network service for the global information when detecting that the period of validity of the current version information has expired.

System and method for acquiring, processing, and updating global information

Partitions of a hosted computing service may be maintained on a computing node. Processing of requests to access the partition may be limited to constrain capacity utilization to a provisioned amount of capacity reserved for the partition. A second, additional amount of capacity may be associated with the partition and may reflect potential future changes to the provisioned amount of capacity. A sum of provisioned and additional capacities associated with partitions on a computing node may be calculated. The computing node may be ranked, relative to other computing nodes, for maintaining new or relocated partitions based on the sum.

Shadowed throughput provisioning

A system, apparatus, and method for maintaining the consistency of global information is disclosed herein. The method includes storing historical version information associated with global information; retrieving current version information associated with the global information from a global information server; generating updated historical version information based on the current version information; transmitting an update confirmation notice to the global information server; and retrieving updated global information from the global information server based on the updated historical version information.

System and method for acquiring, processing and updating global information

The present disclosure is directed to a cross-regional data transmission method and device. Metadata information sent by a corresponding client device is acquired, and the metadata information is sent to a corresponding metadata transmission node device, wherein the metadata information includes data replication progress information of the corresponding client device; to-be-forwarded data information sent by the corresponding metadata transmission node device is acquired, and the to-be-forwarded data information is pushed to the corresponding client device, wherein the to-be-forwarded data information includes data replication progress information of another client device, so that the metadata information sent by the corresponding client device is effectively sent to the corresponding metadata transmission node device; and the acquired to-be-forwarded data information sent by the corresponding metadata transmission node device is pushed to the corresponding client device, to achieve synchronization of the metadata information.

Method of cross-regional data transmission and system thereof

Growing community of inventors

Kenmore, WA, United States of America

Yijun Lu