Emc Corporation

Emc IP Holding Company LLC

Dell Products Inc.

Michael J Scharland

Arieh Don

Patrick Brian Riordan

Rong Yu

Owen Martin

Alexandr Veprinsky

Kenneth A Halligan

David Joshua Brown

Jeremy O'Hare

Ian Wigmore

Steven T McClure

Kuolin Hua

Ofer Michael

Mark J Halstead

Peng Wu

Helen S Raizen

Stephen Richard Ives

Gabriel Benhanokh

Andrew Chanler

Zvi Gabriel Benhanokh

Ran Margalit

Douglas Eric LeCrone

Adnan Sahin

Ramesh Doddaiah

David Lionel Black

Benjamin W Yoder

Kunxiu Gao

Jason J Duquette

Jiahui Wang

James Marriott Guyer

John F Madden, Jr

Jerome Cartmell

Robert DeCrescenzo

Stephen M Lathrop

Pei-Ching Hwang

Violet S Beckett

Assaf Natanzon

David Meiri

Marik Marshak

Parmeshwr Prasad

Erik P Smith

Michael E Bappe

Jonathan I Krasner

Anestis Panidis

Michael Ferrari

James Marriot Guyer

Yechiel Yochai

Philip E Tamer

Sandeep Chandrashekhara

Jaeyoo Jung

Subin George

Peng Yin

Ishay Kedem

Scott Rowlands

Qun Fan

Anoop Raghunathan

Benjamin Randolph

Mohammed Asher

Earl Medeiros

Ningdong Li

Terry Seto Lee

Julie Zhivich

Kevin M Tobin

Kaustubh S Sahasrabudhe

Mark C Lippitt

Jack Fu

Jerome J Cartmell

Vladimir Desyatov

Ro Monserrat

Jeffrey R Nelson

William P Powers, Jr

Daohong Wang

Peter Linden

Thomas Mackintosh

James P McCarthy

Leonard R Linkens

Richard Goodwill

Sean Hu

In a drive cluster in which multiple parity EC (N+P) is implemented, rebuild-related reads are balanced across drives for recovery from drive failure. N members per protection group are read and (P−1) members are skipped, where skipping a member means omission from member rebuild calculations. Per-disk skip counts are calculated, and members that are eligible to be skipped are selected such that per-disk read counts are balanced.

Load balancing for erasure coding with multiple parities

One or more aspects of the present disclosure relate to autonomous write fencing for backup power. In embodiments, an input/output (IO) workload is received at a storage array. Additionally, one or more IO write requests in the IO workload are cached based on a charge level of a backup power supply of the storage array. Further, the cached IO write requests are processed.

Dataset integrity protection

A method, comprising: obtaining a parsed trace log that is generated by parsing a raw trace log of a storage system, the parsed trace log identifying a first sequence of tasks that are executed as part of a same input-output (I/O) operation; classifying the first sequence of tasks by using a machine learning model, the first sequence of tasks being classified into one of a first category and a second category, the first category corresponding to anomalous events, and the second category corresponding to non-anomalous events; when the first sequence of tasks is classified in the first category: identifying an anomaly code that corresponds to the first sequence of tasks and generating an alert that is associated with the anomaly code, the anomaly code being identified based on a distance between the first sequence of tasks and a second sequence of tasks that is mapped to the anomaly code.

I/O path anomaly detection

A method comprising: detecting a storage device failure in a storage array, the storage array including a plurality of storage devices, the storage device failure being a failure of one or more of the plurality of storage devices; wherein: the plurality of storage devices are configured to implement a supergroup including a plurality of subgroups; identifying ones of the plurality of subgroups that are affected by the storage device failure; and rebuilding the identified subgroups, wherein: (A) when a count of failed storage device is less than a first threshold, each of the identified subgroups is rebuilt based only on subgroup parity components for the subgroup without the use of supergroup parity components, and (B) when the count of failed storage devices is between the first threshold and the second threshold, each of the identified subgroups is rebuilt based on both subgroup and supergroup parity components.

Hierarchical parity protection groups for efficient data recovery

In a storage system in which processor cores are exclusively allocated to run process threads of individual emulations, the allocations of cores to emulations are dynamically reconfigured based on forecasted workload. A workload configuration model is created by testing different core allocation permutations with different workloads. The best performing permutations are stored in the model as workload configurations. The workload configurations are characterized by counts of tasks required to service the workloads. Actual task counts are monitored during normal operation and used to forecast changes in actual task counts. The forecasted task counts are compared with the task counts of the workload configurations of the model to select the best match. Allocation of cores is reconfigured to the best match workload configuration.

Autonomous core affinity

One or more aspects of the present disclosure relate to extending the endurance of flash media. In embodiments, an input/output (IO) workload can be received at a storage array. In addition, one or more characteristics of each write-IO in the IO workload can be determined. Further, for each write-IO, a portion of a storage device of a plurality of storage devices can be dynamically allocated based on the one or more characteristics and an endurance level of each of the plurality of storage devices.

Extending flash media endurance

A storage system is configured to accept subsequent versions of write data on a given track to multiple respective slots of shared global memory. A track index table presents metadata at the track level, and can hold up to N slots of data. All slots of shared global memory holding data owed to the source volume and to snapshots of the source volume are bound to the track in the track index table. Each time a write occurs on a track, the track index table is used to determine when a write pending slot for the track is owed to a snapshot copy of the storage volume. When a write pending slot contains data that is owed to a snapshot copy of the source volume, a new slot is allocated to the write IO and bound to the track in the track index table.

Multiple version write pending support in shared global memory

A host IO devices directly implements host read operations on both local memory, and on peer memory via a PCIe non-transparent bridge. When a host read operation is received by a host IO device from a host, the host IO device uses an API to obtain the physical address of the requested data on the peer memory, and generates a PCIe Transaction Layer Packet (TLP) addressed to the address in the peer memory. The TLP addressed to an address in the peer memory is passed over the NTB to the peer compute node to retrieve the data stored in the addressed slot of peer memory. The requested data is returned to the host IO device over the NTB, stored in a buffer, and read out to the host to directy respond to the host read operation.

Host IO device direct read operations on peer memory over a PCIe non-transparent bridge

The meta data containing count and key fields of CKD records are reversibly decoupled from the user data of the data field so that the data can be deduplicated. Multiple CKD records may be coalesced into a larger size CKD track. The coalesced meta data is compressed and stored in a CKD hash table. The user data is hashed, and the hash is used as a hash key that is associated with the compressed meta data in the CKD hash table. When the hash of user data associated with a CKD write IO matches the hash key of an existing entry in the table, data duplication is indicated. The compressed meta data is added to the entry and the user data is deduplicated by creating storage system meta data that points to the pre-existing copy of the user data. The storage system metadata includes unique information that enables the corresponding compressed metadata to be subsequently located in the hash table to reassemble the CKD records.

Inline deduplication for CKD using hash table for CKD track meta data

A method, computer program product, and computer system for receiving, by a computing device, an I/O request. It may be identified whether the I/O request is eligible for handling via a first path without also requiring handling via a second path. If the I/O request is eligible, the I/O request may be processed via the first path on a host I/O stack without processing the I/O request via the second path on a storage array I/O stack. If the I/O request is ineligible, the I/O request may be processed via the first path on the host.

System and method for accelerated I/O access using storage array driver in Off-the-shelf server

Aspects of the present disclosure relate to data cache management. In embodiments, a logical block address (LBA) bucket is established with at least one logical LBA group. Additionally, at least one LBA group is associated with two or more distinctly sized cache slots based on an input/output (IO) workload received by the storage array. Further, the association includes binding the two or more distinctly sized cache slots with at least one LBA group and mapping the bound distinctly sized cache slots in a searchable data structure. Furthermore, the searchable data structure identifies relationships between slot pointers and key metadata.

Cache memory architecture and management

The present disclosure relates to one or more memory management techniques. In embodiments, one or more regions of storage class memory (SCM) of a storage array is provisioned as expanded global memory. The one or more regions can correspond to SCM persistent cache memory regions. The storage array's global memory and expanded global memory can be used to execute one or more storage-related services connected to servicing (e.g., executing) an input/output (IO) operation.

Tiered persistent memory allocation

Aspects of the present disclosure relate to data cache management. In embodiments, a storage array's memory is provisioned with cache memory, wherein the cache memory includes one or more sets of distinctly sized cache slots. Additionally, a logical storage volume (LSV) is established with at least one logical block address (LBA) group. Further, at least one of the LSV's LBA groups is associated with two or more distinctly sized cache slots based on an input/output (IO) workload received by the storage array.

Embodiments of the present disclosure relate to cache memory management. One or more global caches are dynamically partitioned and sized into one or more cache partitions based on anticipated input/output (IO) workloads.

Cache memory management

When a read request for the data portion is received from an application executing on a host, the host may determine whether the data portion is in host cache, and if so, whether the logical storage unit of the data portion is shared by another host system. If there is another host system sharing the logical storage unit, a latest version stored on the storage system may be determined and compared to the version stored in the host cache. If the version in the host cache is the same as the latest version stored on the storage system, the data portion may be retrieved from the host cache. If the version in the host cache is not the latest version stored on the storage system, the data portion may be retrieved from the storage system, and the host cache may be updated with the latest version of the data portion.

Host cache coherency when reading data

Growing community of inventors

Franklin, MA, United States of America

Michael J Scharland