Microsoft Technology Licensing, LLC

James Robert Benton

Ran Kalach

Paul Adrian Oltean

Abhishek Gupta

Ronakkumar N Desai

Chun Ho Cheung

Jin Li

Neal Robert Christiansen

Sarosh Cyrus Havewala

Sudipta Sengupta

Ravinder S Thind

Jason Daniel Shay

Michael S Murstein

Ahmed El-Shimi

Kashif Hasan

Georgi M Matev

Curt Andrew Steeb

Mukesh Karki

Rajeev Y Nagar

Paul C Sutton

Kartik Nadipuram Raghavan

Raymond D Pedrizetti

Spencer Alan Bishop

Ivelin Lyubenov Zheglov

Cristian Teodorescu

Joerg-Thomas Pfenning

W Jeff Westerinen

Thomas C Phillips

Chun Ho (Ian) Cheung

Alan R Morin

Disclosed herein is a system and method to synchronize files from a hosted file system where the client endpoint devices have direct access to the files stored on a hosted file store. When the client desires to synchronize with the files stored on the hosted file store the process identifies a sync role that can process the request. The particular sync role does not need to maintain state with the client device or the file store. The state necessary for the processing of the synchronization process is maintained separate from the sync role such that any sync role in the system can pick up the requests and process the requests. Once the role has been selected the synchronization process proceeds as normal. As state is not maintained by the sync role multiple roles can process the request or two roles can handle different portions of the process.

Hosted file sync with stateless sync nodes

Data streams may be stored in a chunk store in the form of stream maps and data chunks. Data chunks corresponding to a data stream may be stored in a chunk container, and a stream map corresponding to the data stream may point to the data chunks in the chunk container. Multiple stream maps may be stored in a stream container, and may point to the data chunks in the chunk container in a manner that duplicate data chunks are not present. Techniques are provided herein for localizing the storage of related data chunks in such chunk containers, for locating data chunks stored in chunk containers, for storing data streams in chunk stores in localized manners that enhance locality and decrease defragmentation, and for reorganizing stored data streams in chunks stores.

Scalable chunk store for data deduplication

Disclosed herein is a system and method that can be used to synchronize files from a hosted file system where the client endpoint devices have direct access to the files stored on a hosted file store/system. When the client synchronizes with the files stored on the hosted file store the synchronization process identifies the files that need to be synchronized with the client and prepares batches of files to be synced. During the process of synchronization the files are not locked from other user's being able to use the files. If the file is modified by another endpoint during the file synchronization process the file will not be synchronized with the client device during the particular synchronization process. In this way synchronization becomes possible with the hosted service without impacting the performance of other devices that use and access the data in the hosted file store.

Hosted file sync with direct access to hosted files

The subject disclosure is directed towards a data deduplication technology in which a hash index service's index is partitioned into subspace indexes, with less than the entire hash index service's index cached to save memory. The subspace index is accessed to determine whether a data chunk already exists or needs to be indexed and stored. The index may be divided into subspaces based on criteria associated with the data to index, such as file type, data type, time of last usage, and so on. Also described is subspace reconciliation, in which duplicate entries in subspaces are detected so as to remove entries and chunks from the deduplication system. Subspace reconciliation may be performed at off-peak time, when more system resources are available, and may be interrupted if resources are needed. Subspaces to reconcile may be based on similarity, including via similarity of signatures that each compactly represents the subspace's hashes.

Using index partitioning and reconciliation for data deduplication

The subject disclosure is directed towards a data storage service that uses hash values, such as substantially collision-free hash values, to maintain data integrity. These hash values are persisted in the form of mappings corresponding to data blocks in one or more data stores. If a data error is detected, these mappings allow the data storage service to search the one or more data stores for data blocks having matching hash values. If a data block is found that corresponds to a hash value for a corrupted or lost data block, the data storage service uses that data block to repair the corrupted or lost data block.

Data error detection and correction using hash values

The subject disclosure is directed towards transforming a file having at least one undeduplicated portion into a fully deduplicated file. For each of the at least one undeduplicated portion, a deduplication mechanism defines at least one chunk between file offsets associated with the at least one undeduplicated portion. Chunk boundaries associated with the at least one chunk are stored within deduplication metadata. The deduplication mechanism aligns the at least one chunk with chunk boundaries of at least one deduplicated portion of the file. Then, the at least one chunk is committed to a chunk store.

Optimization of a partially deduplicated file

The subject disclosure is directed towards changing a file from a fully deduplicated state to a partially deduplicated state in which some of the file data is deduplicated in a chunk store, and some is recalled into the file, that is, in the file's storage volume. A partial recall mechanism such as in a file system filter tracks (e.g., via a bitmap in a file reparse point) whether file data is maintained in the chunk store or has been recalled to the file. Data is recalled from the chunk store as needed, and committed (e.g., flushed) to the file. Also described is efficiently returning the file to a fully deduplicated state by using the tracking information to determine which parts of the file are already deduplicated into the chunk store so as to avoid their further deduplication processing.

Partial recall of deduplicated files

The subject disclosure is directed towards data deduplication (optimization) performed by phases/modules of a modular data deduplication pipeline. At each phase, the pipeline allows modules to be replaced, selected or extended, e.g., different algorithms can be used for chunking or compression based upon the type of data being processed. The pipeline facilitates secure data processing, batch processing, and parallel processing. The pipeline is tunable based upon feedback, e.g., by selecting modules to increase deduplication quality, performance and/or throughput. Also described is selecting, filtering, ranking, sorting and/or grouping the files to deduplicate, e.g., based upon properties and/or statistical properties of the files and/or a file dataset and/or internal or external feedback.

Extensible pipeline for data deduplication

Method and system for establishing and maintaining quotas. An auto quota is defined and applied to a directory. Input and output is monitored to detect a successful operation that involves a subdirectory of the directory. A determination is made as to whether to apply a quota associated with the auto quota to the subdirectory. If the determination is that the quota is to be applied to the subdirectory, it is automatically applied.

Auto quota

Method and system for maintaining namespace consistency between selected objects maintained by a file system and a filter associated therewith. Metadata regarding selected objects of a file system is maintained by a filter while the filter is attached to the file system and persisted in non-volatile storage. The namespace of the file system may be changed while the filter is unattached from the file system. Afterwards, when the filter is attached to the file system, the namespace of the filter is synchronized with the namespace of the file system for the selected objects.

Method and system of detecting file system namespace changes and restoring consistency

Described is a method and system by which reports of storage usage in computer systems are generated in an efficient manner by consolidating multiple requests for reports into a minimal number of volume scans, including by intelligently selecting a scanning method (e.g., of file system metadata versus find-first/find-next) and by performing parallel scans on different volumes. Namespace consolidation scans namespaces together, so as to generate multiple reports from the same set of files, reducing the number of volumes scans required to collect the data. Each volume scan may be a find-first, find next directory-based scan, or a volume metadata database scan. Time consolidation groups independent storage report generations together, such as storage report requests received within an administrator-specified interval. Parallel scans of different volumes may be performed, subject to I/O and processing resource limitations, and so that volumes partitioned on the same spindle are not scanned in parallel.

Method and system for efficient generation of storage reports

Described is a method and system by which storage reports are generated from a volume snapshot set, rather than from a live volume. A volume snapshot set includes a representation or copy of a volume at a single point in time. By scanning the snapshot, a consistent file system image is obtained. Scanning may take place by enumerating a volume's directories of files, or, when available, by accessing a file system metadata of file information (e.g., a master file table) separately maintained on the volume. With some (e.g., hardware-based) snapshot technologies, the snapshot can be transported to another computing system for scanning by that other computing system, thereby avoiding burdening a live system's resources when scanning. Accurate and consistent storage reports are thus obtained at a single point in time, independent of the number of volumes being scanned.

Generating storage reports using volume snapshots

Described is a storage reports duplicate file detector that operates by receiving file records during a first scan of file system metadata. The detector computes a hash based on attributes in the record, and maintains the hash value in association with information that indicates whether a hash value corresponds to more than one file. In one implementation, the information corresponds to the amount of space wasted by duplication. The information is used to determine which hash values correspond to groups of potentially duplicate files, and eliminate non-duplicates. A second scan locates file information for each of the potentially duplicate files, and the file information is then used to determine which groups of potentially duplicate files are actually duplicate files.

Storage reports duplicate file detection

Described is a storage reports scanner that works to generate reports of storage usage in computer systems in an efficient manner. The scanner receives a set of namespaces for a file system volume from a storage reports engine. The scanner scans file system metadata to construct a directory table of entries corresponding to a directory tree of nodes representative of the hierarchy of directories of the file system volume. Each node corresponding to a namespace in the namespace set is marked as included. A second scan of the file system metadata determines, for each file, whether that file is in or under an included directory by accessing the directory table. For each file that is in or is under an included directory, file information is returned to the engine. The engine may request the scanner to provide full path information, which the scanner determines via the directory table.

Growing community of inventors

Seattle, WA, United States of America

James Robert Benton