Microsoft Technology Licensing, LLC

Emc IP Holding Company LLC

Emc Corporation

Other

Aivot, LLC

Shashwat Srivastav

Bradley Gene Calder

Niranjan Nilakantan

Chen Xiao Wang

Sriram Sankaran

Jun Luo

Shane Kumar Mainali

Jiesheng Wu

Padmanabha Chakravarthy Uddaraju

Abdul Rafay Abbasi

Vishrut Shah

Qi Zhang

Ju Wang

Huapeng Yuan

Andrew Robertson

Peter Musial

Subba R Gaddamadugu

Arild Einar Skjolsvold

Jie Song

Hemal Khatri

Wei Yin

Mikhail Danilov

Ivan Tchoub

Deepali Bhardwaj

Mikhail Malygin

Andrey Fomin

Xinran Wu

Harshawardhan Gadgil

Stephen Glen Graham

Maxim S Trusov

Maxim Mazeev

Alexander Felsöbüki Nagy

Leonidas Rigas

Karthik Navaneethakrishnan

Lu Lei

Dong Wang

Ao Sun

Fredrick Allen Crable

David Gillam

Alexander Rakulenko

Igor Medvedev

Yu Teng

Ronnie Yu Cai

Matthew L Troutman

Jialei Wu

Brian D Burck

Alexander Felsobuki Nagy

Liang Mei

Daquan Zuo

Embodiments of the present invention relate to synchronously replicating data in a distributed computing environment. To achieve synchronous replication both an eventual consistency approach and a strong consistency approach are contemplated. Received data may be written to a log of a primary data store for eventual committal. The data may then be annotated with a record, such as a unique identifier, which facilitates the replay of the data at a secondary data store. Upon receiving an acknowledgment that the secondary data store has written the data to a log, the primary data store may commit the data and communicate an acknowledgment of success back to the client. In a strong consistency approach, the primary data store may wait to send an acknowledgement of success to the client until it receives an acknowledgment that the secondary has not only written, but also committed, the data.

Synchronous replication in a distributed storage environment

Presented herein are methods of replicating versioned and hierarchical data structures, as well as data structures representing complex transactions. Due to interdependencies between data entities and a lack of guaranteed message ordering, simple replication methods employed for simple data types cannot be used. Operations on data structures exhibit dependencies between the messages making up the operations. This strategy can be extended to various types of complex transactions by considering certain messages to depend on other messages or on the existence of other entries at the data store. Regardless of origin, these dependencies can be enforced by suspending the processing of messages with unsatisfied dependencies until all of its dependencies have been met. Alternately, transactions can be committed immediately, creating entities that include versioned identifiers for each of their dependencies. These entities can then be garbage collected of the parent objects are not subsequently created.

Versioned and hierarchical data structures and distributed transactions

The present disclosure generally relates to performance recreation, and in particular, the recreation of observed human performance using reinforcement learning. In this regard, a first object is identified from a plurality of objects. The manipulation of the first object is tracked from a first position to a second position. A characterization of the manipulation is generated. A policy that controls a mechanical gripper to recreate the manipulation is generated based on an iteratively increasing cumulative award. The mechanical gripper iteratively recreates the manipulation to increase a cumulative award with each recreation.

Performance recreation system

A method, system, and computer program product for providing, via a provisioning engine, a scalable set of indexed key-value pairs enabled to store objects in a data storage environment; wherein the data representing the objects is enabled to be spread across arrays in the data storage environment; wherein additional arrays are enabled to be added to the data storage environment and included in the indexed key-value pairs; wherein the data stored across the arrays may be balanced.

Scalable index store

An efficient and scalable method of managing data protection in a geographically distributed storage system is provided using erasure coding. Each zone in the geographically distributed storage system of multiple zones or clusters encodes primary backup chunks replicated from original data chunks in remote zones to create coding chunks. The zone retains one of the coding chunks and distributes the remaining coding chunks to remote zones for future recovery from dual and more complex failures. Upon completion, the primary backup chunks and peer backup chunks can safely be deleted to reduce overhead. Encoding k primary backup chunks to create m coding chunks, and distributing the m coding chunks to different remote zones assures recovery from 1 to m zone/cluster/chunk failures. After 1 to m zone failures the unaffected zones are each able to recover all of the unavailable chunks by retrieving from each other any of the unaffected zone's distributed coding chunks and still available original data chunks, and restoring the content of the unavailable chunks by decoding the combination of the retained coding chunk, distributed coding chunk, and any still available original data chunks.

Data protection management for distributed storage

A computer implemented method, system, and computer program product for enabling strong consistency of unique objects between zones comprising enabling consistency of an unique data object between in zones by switching between consistency techniques to keep the unique data object consistent between the zones; wherein read and write access is enabled to the unique object in the zones.

Zone consistency

Structures and processes for garbage collection of search trees under Multi-Version Concurrency Control (MVCC). Such search trees may be used to store data within a distributed storage system. A process detects live search tree elements using tracing and then identify storage chunks having no live elements as garbage to be reclaimed. The process can be paused and resumed to reduce impact on other system processing.

Tracing garbage collector for search trees under multi-version concurrency control

A computer-executable method, computer program product and system for managing metadata in a distributed data storage system, wherein the distributed data storage system includes a first node and one or more data storage arrays, the computer-executable method, computer program product and system comprising partitioning management of metadata created in the distributed data storage system into one or more portions of metadata, wherein the first node manages a first portion of the one or more portions of metadata, and storing the metadata using the first node.

Distributed metadata management

A computer program product, system, and method for generating and storing a parity chunk using at least a first chunk and a second chunk; detecting at least one unused segment within the first chunk; generating a new chunk as a third chunk; associating one or more used segments from the first chunk to the third chunk; filling an unused segment within the third chunk with new data; calculating a delta parity between the first chunk and the third chunk; and updating the stored parity chunk using the delta parity.

Efficient garbage collection for distributed storage with forward error correction

Structures and processes for garbage collection of search trees under Multi-Version Concurrency Control (MVCC). Such search trees may be used to store data within a distributed storage system. A process detects live search tree elements using tracing and then identify storage chunks having no live elements as garbage to be reclaimed. The process can be paused and resumed to reduce impact on other system processing. To reduce disk fragmentation, a garbage collector may copy pages between chunks prior to reclaiming chunk capacity. Also described is a resource efficient scheduler for a garbage collection.

Copying garbage collector for B+ trees under multi-version concurrency control

A computer-executable method, computer program product, and system for managing backups in a distributed data storage system including a first zone, a second zone, and a third zone, the computer-executable method, computer program product, and system comprising processing, at the third zone, a first portion of data of the first zone and a second portion of data of the second zone to create a combined portion of data, and removing the first portion and second portion from the third zone.

Data backup management on distributed storage systems

Systems and methods for determining garbage collection (GC) scope in a distribute storage system using chunk-based storage. The systems and methods are compatible with multi-version concurrency control (MVCC) semantics.

Garbage collection scope detection for distributed storage

Partition management for a scalable, structured storage system is provided. The storage system provides storage represented by one or more tables, each of which includes rows that represent data entities. A table is partitioned into a number of partitions, each partition including a contiguous range of rows. The partitions are served by table servers and managed by a table master. Load distribution information for the table servers and partitions is tracked, and the table master determines to split and/or merge partitions based on the load distribution information.

Partition management in a partitioned, scalable, and available structured storage

Growing community of inventors

Seattle, WA, United States of America

Shashwat Srivastav