Qumulo, Inc.

Thomas Gregory Rothschilds

Yuxi Bai

Jonathan Michael MacLaren

Aaron Passey

Michael Anthony Chmiel

Noah Trent Nelson

Matthew Christopher McMullan

Tyler Morrison Moody

Pathirat Kosakanchit

Kimberly Alice Fairhurst

Rowan Arthur Phipps

Peter John Godman

Patrick Jakubowski

Edward Carpenter

Thomas R Unger

Steven Henry Haber

Remi Bernotavicius

Duncan Robert Fairbanks

Edward Brow

Eric Erwin Youngblut

William Ehlhardt

Ying Fairweather

Zexuan Wang

Conner Saltiel Hansen

Hanqing Zhang

David Patrick Rogers, Iii

Graham Edwin Ellis

Thorne Davis Garvin

Neal Thomas Fachan

Stephen Craig Fleischman

Porter Michael Smith

Jeremy Drouillard

Embodiments are directed to a file system that include object stores. An object store for write requests may be provided. Write ahead log (WAL) entries that include data blocks may be generated. A WAL object may be generated based on the WAL entries and stored in the object store. An in-memory overlay may be updated to associate the data blocks with the WAL object. A checkpoint operation may be executed to: generate an index object that includes index entries that associate other data blocks with data objects stored in the object store; update the index object to include index entries that associate the data blocks with the WAL object; store the updated index object in the object store; update the in-memory overlay to remove the association of the data blocks and the WAL object and update the in memory WAL to remove records of successfully checkpointed WAL objects.

Object-based storage with garbage collection and data consolidation

Embodiments are directed to managing data in a file system. In response to a request from a client of a spoke file system to access file system items shared by a hub file system: determining a portion of the spoke file systems that may have an existing lock on the file system items; obtaining log entries from the portion of the spoke file systems such that the log entries are associated with commands that were executed in the portion of the spoke file systems; executing the commands in the hub file system to update the file system items in the hub file system; granting a lock on the file system items in the hub file system that enables the client to access the updated file system items from the spoke file systems.

Global namespaces for distributed file systems

Embodiments are directed to tiered data store with persistent layers. A write tier in the file system for storing in a file system. A value for a performance metric that corresponds to write requests to the file system may be predicted based on characteristics of the write requests such that the performance metric may be determined based on a plurality of interactions with the write tier. The predicted value that exceeds a threshold value of the performance metric may be employed to cause performance of further actions, including: queuing a portion of the write requests in a memory buffer based on the predicted value and the threshold value; combining the queued portion of the write requests into s; storing the data segments in the write tier such that a measured value of the performance metric may be less than the threshold value.

Tiered data storage with ephemeral and persistent tiers

Embodiments are directed to managing file systems over a network. A core specification that defines characteristics of a portion of a file system and parameters may be provided. The characteristics may include one or more storage device characteristics. And, the parameters may include one or more of a cluster size, or a protection level. File system models may be generated based on the core specification and the parameters. Each file system model may correspond to a file system that conforms to a core specification and specific values for the parameters. A simulation engine may be employed to provide simulation results based on the file system models. Deployment models may be generated based on the simulation results. The deployment models may be employed to provision file systems that conform to the file system performance characteristics associated with its corresponding deployment model.

Predictive performance analysis for file systems

Embodiments are directed to error resistant logging. A write-ahead log (WAL) for preserving a verifiable record of file system activity may be provided such that the WAL may include a sequence of log blocks that each may include a cyclic redundancy check (CRC) value of a next adjacent log block in the WAL. In response to executing a file system operation associated with payload data, further actions may be performed, including: generating log blocks based on the payload data; generating a log segment that includes the log blocks; modifying a portion of the contents of a head-block of the log segment such that the modified head-block of the log segment has a CRC value that matches a CRC value of a tail-block of the WAL; copying the contents of the modified head-block of the log segment into the tail-block of the WAL.

Error resistant write-ahead log

A file system with distributed resource caching that includes cache volumes and agents that may be associated with clients of the file system may be provided. A cache allocation for each agent may be determined based on a capacity of the cache volumes and a number of the agents such that each cache allocation is associated with tokens that each represents a reserved portion of free space in the cache volumes. Storage jobs may be provided to the agents. Data associated with the storage jobs may be stored in the cache volumes. The cache allocation for each agent may be reduced based on the data stored for each agent.

Distributed resource caching by reallocation of storage caching using tokens and agents with non-depleted cache allocations

Embodiments are directed to a file system engine that provides a file system that includes a plurality of blocks on a file storage tier with portion of the blocks associated with a cache storage tier. A cache engine provides heat extents that include a heat score and a run length value such that the heat extents are represent activities associated with the blocks. Heat extents associated with file system activities are determined based on a position of the blocks that corresponds with the run length values of the heat extents and sectors on the file storage tier. Heat scores associated with the heat extents may be modified based on the activities and distribution models to conform the heat extents with the distribution models. Blocks associated with heat scores that are less than a threshold value provided by the distribution models may be disassociated from the cache storage tier.

File system cache tiers

Embodiments are directed to managing file systems over a network. Jobs may be provided to a storage computer in a file system. Control models may be associated with the jobs. Scores may be generated based on the control models. Each job may be associated with a score provided by its associated control model. And, each job that may be behind its corresponding schedule may be associated with a higher score value than each other job that may be either on its corresponding other schedule or ahead of its corresponding other schedule. Commands may be selected for execution based on the commands being associated with a job that may be associated with the higher score value that may be greater than score values associated with other jobs. The jobs may be ranked based on the updated scores. Subsequent commands may be selected and executed based on the ranking of the jobs.

Managing throughput fairness and quality of service in file systems

A facility comprising systems and method for automatically triggering the collection of comprehensive monitoring information in a distributed processing system. The facility compares the overall performance of distributed processing system to one or more performance metrics and, in response to determining that one or more performance metrics is not satisfied, triggers one or more of the nodes within the distributed processing system to increase one or more of its monitoring rate or its distribution rate. The facility collects and analyzes the collected information to provide resources that can be used to assess and diagnose failures within the distributed processing system. In this manner, the facility reacts to performance anomalies by triggering nodes within in the system to provide comprehensive performance information over a trigger period for diagnostic purposes.

Triggering the increased collection and distribution of monitoring information in a distributed processing system

Embodiments are directed to managing data in a file system. A pre-fetch engine may receive requests from a client the file system that includes a pre-fetch storage tier and a file storage tier of storage devices. The pre-fetch engine determines a pre-fetch policy based on the requests such that the pre-fetch policy determines which blocks to copy to the pre-fetch storage tier. And, the pre-fetch policy may be associated with a score model that includes score rules where one of the rules may be associated with a client score. The pre-fetch engine may obtain scores associated with the score rules such that the scores are based on previous requests made by the client. In response to scores exceeding a threshold value, the pre-fetch engine may copy the blocks to the pre-fetch storage tier. The pre-fetch engine may update the scores based on the performance of the pre-fetch policy.

Client aware pre-fetch policy scoring system

Growing community of inventors

Seattle, WA, United States of America

Thomas Gregory Rothschilds