International Business Machines Corporation

International Business Corporation

Globalfoundries Inc.

Machines Corporation

Lokesh M Gupta

Kevin J Ash

Matthew G Borlick

Michael T Benhase

Kyler A Anderson

Matthew J Kalos

Beth Ann Peterson

Karl Allen Nielsen

Trung Ngoc Nguyen

Clint Alan Hardy

Brian Anthony Rinaldi

Carol Santich Mellgren

Micah Robison

Kenneth Wayne Todd

Theresa Mary Brown

Binny Sher Gill

Sonny Earl Williams

Evangelos Stavros Eleftheriou

Roman Alexander Pletka

Robert Haas

Brent Cameron Beardsley

Ioannis Koltsidas

Roger Gregory Hathorn

Gail Andrea Spear

Joseph S Hyde, Ii

Stephen LaRoux Blinick

Xiao-Yu Hu

Brian D Hatfield

Matthew Richard Craig

Cheng-Chung Song

John Charles Elliott

Warren Keith Stanley

David Blair Whitworth

Alfred Emilio Sanchez

Theodore Timothy Harris, Jr

Will A Wright

Joseph Hayward

Yu-Cheng Hsu

Brian James Cagno

Chung Man Fung

Lee Charles LaFrese

David Brent Schreiber

Adrian Cuenin Gerhard

Suguang Li

Michael Thomas Benhase, Jr

Matthew Sanchez

James Lee Hafner

Susan Kay Candelaria

John Glenn Thompson

Nicolas Marc Clayton

Kurt Allen Lovrien

Mark Elliott Hack

Jared Michael Minch

Ellen Joanne Grusy

Edward Hsiu-Wei Lin

Christopher J Strauss

Xiao Y Hu

Paul Henri Muench

Raul Eduardo Saba

Louis Alonso Rasor

James E Olson

Richard P Oubre, Jr

Usman Ahmed

Gregory E McBride

Todd Charles Sorenson

Rivka Mayraz Matosevich

Roger C Raphael

Nedlaya Yazzie Francisco

Edward Lin

Steven Robert Lowe

Richard Harry Hopkins

Boyan Zhao

Keneth W Todd

Matthew John Ward

Joshua J Crawford

Lawrence Yiumchee Chiu

Paul Nicholas Cashman

Lawrence Carter Blount

Mark L Lipets

Christopher Daniel Filachek

Michael John Jones

Juan Antonio Yanes

Seamus Burke

Xiaoyu Hu

Sean Patrick Riley

Mark A Lehrer

Markus Navarro

Donald Prentiss Terry

Alistair Leask Symon

Rose L Manz

Carol Santich Melgren

Kenney Nian Gan Giu

Richard H Hopkins

Kyle A Anderson

An example method for dynamic data relocation using cloud based ranks comprises monitoring accesses to data stored on a plurality of local ranks of an enterprise storage system; identifying data which has not been accessed for a predetermined amount of time based on the monitored accesses; and moving the data which has not been accessed for the predetermined amount of time to one or more cloud based ranks of the enterprise storage system, wherein each cloud based rank comprises storage space on one or more cloud storage devices, the storage space on the one or more cloud storage devices mapped to corresponding virtual local addresses that are grouped as a virtual local rank.

Dynamic data relocation using cloud based ranks

Described are techniques for selecting compressed or non-compressed data storage using hints for a dataset. The techniques include receiving a dataset at a data storage system that manages storage location on compressed data storage and non-compressed data storage. The techniques further include determining, by the data storage system, that the dataset includes one or more hints for selecting a storage location for the dataset from one of the storage locations managed by the data storage system. The techniques further include selecting, by the data storage system, the storage location for the dataset based in part on the one or more hints and storing the dataset to the storage location.

Hinting information for selecting a storage location for a dataset

Provided are a computer program product, system, and method for determining a storage device type on which to store extents of data in a storage system. The storage system provides a first storage device type that has native data compression at a first compression ratio and a second storage device type that has native data compression at a second compression ratio. The second compression ratio is higher than the first compression ratio. A determination is made for an extent to store the extent on the first storage device type or the second storage device type based on access frequency to data in the extent and an Input/Output (I/O) priority level for the extent. The extent is stored on the determined first or second storage device type.

Determining a storage device type on which to store extents of data in a storage system

A computational device receives an indication that specifies a maximum retention time in cache for a first plurality of tracks, wherein no maximum retention time is specified for a second plurality of tracks. A plurality of insertions points are generated in a least recently used (LRU) list, wherein different insertion points in the LRU list correspond to different amounts of time that a track of the first plurality of tracks is expected to be retained in the cache, wherein the LRU list is configured to demote both tracks of the first plurality of tracks and the second plurality of tracks from the cache.

Integration of application indicated maximum time to cache to least recently used track demoting schemes in a cache management system of a storage controller

Provided are a storage device, system, and method for performance enhancing measures for storage in which high-performance volumes reside. Indication is made of a volume configured in the arrays to have high-performance. An array is determined storing the indicated volume. A determination is made of storage devices in which the determined array resides. The storage devices are enabled to perform garbage collection on blocks of data having valid and invalid pages in the storage devices at a first garbage collection rate and a second garbage collection rate. The second garbage collection rate causes a storage device to perform garbage collection with respect to a greater number of blocks than the first garbage collection rate. A command is sent to the determined storage devices to garbage collect at the second garbage collection rate.

Garbage collection for storage in which high-performance volumes reside

Provided are a computer program product, system, and method for generating data protection directives to provide to a storage controller to control access to data in cache. A data protection directive is generated for a data subset indicating access request type and a protective action with respect to the access request type for the data subset. The data protection directive is transmitted to the storage controller. The storage controller includes the data protection directive in metadata for the data subset. The data protection directive causes the storage controller to perform the protective action in response to an access request of the access request type to a portion of the data subset.

Generating data protection directives to provide to a storage controller to control access to data in cache

A machine learning module is trained by receiving inputs comprising attributes of a computing environment, where the attributes affect a likelihood of failure in the computing environment. In response to an event occurring in the computing environment, a risk score that indicates a predicted likelihood of failure in the computing environment is generated via forward propagation through a plurality of layers of the machine learning module. A margin of error is calculated based on comparing the generated risk score to an expected risk score, where the expected risk score indicates an expected likelihood of failure in the computing environment corresponding to the event. An adjustment is made of weights of links that interconnect nodes of the plurality of layers via back propagation to reduce the margin of error, to improve the predicted likelihood of failure in the computing environment.

Perform preemptive identification and reduction of risk of failure in computational systems by training a machine learning module

A computer-implemented method, according to one approach, includes: monitoring performance in NVRAM having a plurality of memory blocks. Moreover, a change in the flow rate of I/O requests received at the NVRAM is identified as a result of the monitoring. In response to identifying the change is an increase in the flow rate of received I/O requests, the percentage of invalid pages in a given block of NVRAM that triggers garbage collection to be performed on the given block is increased. However, in response to identifying the change is a decrease in the flow rate of received I/O requests, the percentage of invalid pages in a given block that triggers garbage collection to be performed on the given block is decreased.

Adjusting garbage collection based on input/output flow rates

Provided are a computer program product, system, and method for determining whether to compress data at a source device to transmit to a target device. A determination is made of network bandwidth between the source device and the target device. The source device transmits data to the target device. A determination is made of source available processor resources on the source device and target available processor resources on the target device. A determination is made as to whether to compress data at the source device to transmit to the target device based on the network bandwidth, the source available processor resources, and the target available processor resources. Compression is initiated, on the source device, of the data to transmit to the target device in response to determining that the source device is to compress the data.

Determining whether to compress data at a source device to transmit to a target device

Provided are techniques for task adjustment for uniform performance of drives. Arrays are created for different drive types. Performance benchmarks are run for each of the arrays to compute a stage maximum value and a destage maximum value for each of the drive types. At run time, in response to determining that a mode indictor indicates a uniform performance mode, a maximum number of stage Task Control Blocks (TCBs) are set to a stage maximum uniform value and a maximum number of destage TCBs are set to a destage maximum uniform value. The stage of first data is performed using an actual number of stage TCBs that do not exceed the maximum number of stage TCBs. The destage of second data is performed using an actual number of destage TCBs that do not exceed the maximum number of destage TCBs.

Task adjustment for uniform performance of drives

Provided are a method, a system, and a computer program product in which metadata associated with encrypted data is maintained in a cloud computing environment, where the metadata indicates whether reading of information in the encrypted data is restricted geographically. A controller provides a decryption code to a cloud server located in a geographical location. The decryption code is for decrypting the encrypted data. The controller provides the decryption code, based on a determination as to whether the metadata indicates whether the reading of information in the encrypted data is restricted geographically.

Encryption and decryption of data in a cloud storage based on indications in metadata

Provided are a computer program product, system, and method for using a transient cache list and prolonged cache list to manage tracks in cache based on a demotion hint with a track access request. A track is staged into the cache in response to a request for the track. A determination is made as to whether the request provides a demotion hint. The track is indicated in a transient cache list in response to determining that the request provides the demotion hint. The track is indicated in a prolonged cache list in response to determining that the request does not provide the demotion hint. The track is demoted from the cache in response to the request for the track comprising a read request, returning the track to the read request, and the track is indicated in the transient cache list.

Using a transient cache list and prolonged cache list to manage tracks in cache based on a demotion hint with a track access request

Provided are a computer program product, system, and method for indicating extents of tracks in mirroring queues based on information gathered on tracks in extents in cache. Extent information on an extent of tracks in a cache indicated in an active cache list is processed in response to destaging a track from the active cache list to add to a demote list used to determine tracks to remove from the cache. The extent information is related to a number of modified tracks in an extent destaged from the active cache list. The extent information for the extent is used to determine one of a plurality of mirroring queues to indicate the extent including modified tracks. A mirroring queue having a higher priority than another mirroring queue is processed at a higher rate to determine extents of tracks to mirror from the cache to the secondary storage.

Indicating extents of tracks in mirroring queues based on information gathered on tracks in extents in cache

Provided are a computer program product, system, and method for saving track metadata format information for tracks demoted from cache for use when the demoted track is later staged into cache. A track is demoted from the cache and indicated in a demoted track list. The track format information for the demoted track is saved, wherein the track format information indicates a layout of data in the track. The saved track format information for the demoted track is used when the demoted track is staged back into the cache.

Saving track metadata format information for tracks demoted from cache for use when the demoted track is later staged into cache

Provided are a computer program product, system, and method for determining whether to process a host request by training a machine learning module. A machine learning module receives input information related to at least one of running tasks, mail queue messages related to host request processing, I/O request processing, and a host request received from a host system and produces output for the host request indicating whether to process the host request. A determination is made of output produced from the machine learning module based on the input information for a received host request. A determination is made of an expected output based on a measured system resource usage and an optimum system resource usage. The machine learning module is retrained with the input information, the output produced from the machine learning module, and the expected output and provided to determine whether to process a received host request.

Determining whether to process a host request by training a machine learning module

Growing community of inventors

Tucson, AZ, United States of America

Lokesh M Gupta