International Business Machines Corporation

Thomas K Clark

Jason Christopher Young

James John Seeger, Jr

Venkateswararao Jujjuri

Bernhard Julius Klingenberg

Stephen F Correl

John Jason Auvenshine

Robert Michael Rees

Craig Fulmer Everhart

Neeta Garimella

Ninad S Palsule

Malahal R Naineni

Russell H Weight

James J Seeger

Paul A Dorwin

Andreas J Moran

Ramani Ranjan Routray

Frank Bernhard Schmuck

Joseph W Dain

David Maxwell Cannon

Gregory Richard Hintermeister

Yang Song

Roger C Raphael

Colin Scott Dawson

Soumitra Sarkar

Douglas J Griffith

Yi Zhou

Roger Lee Haskin

Theodore Brian Vojnovich

Michael Allen Kaczmarski

Shah Mohammad Rezaul Islam

Nilesh Prabhakar Bhosale

Felipe Knop

Myung Mun Bae

Robert J Curran

Wayne Curtis Hineman

Toby Lyn Marek

Sudhir Gurunandan Rao

Bruce Light Hillsberg

Michael Wayne Young

Ramakrishna Dwivedula

Mandar Jayant Vaidya

David M Wolf

Storage network with controller that includes using deep packet inspection to determine whether a computer file or computer object is sensitive and/or critical, and then, if it is sensitive or critical, applying an artificial intelligence algorithm to determine whether there is a potential threat that is represented by the file or object. If there is a potential threat, then a cyber-resiliency workflow is performed to eliminate or at least mitigate the potential threat.

Detecting security threats to data on a storage controller

An example operation may include one or more of connecting, by a data owner node, to a blockchain network that includes a plurality of storage provider peer nodes associated with a plurality of cloud storage providers, selecting, by the data owner node, storage provider peer nodes of the plurality of the storage provider peer nodes to be approved storage location member nodes, and executing, by the data owner node, a smart contract to: store data on the cloud storage providers of the plurality of the cloud storage providers associated with the approved storage location member nodes, and record movement of the data to the cloud storage providers on the blockchain.

Tracking data transfers

A root user identifier of a computing system is disabled. Thereafter, and in response to determining that a problem with the computing system requires root privileges to the computing system to solve, a code patch for installation on the computing system is received from a third party. The code patch is installed on the computing system, resulting in a user identifier temporarily having the root privileges to the computing system. The user identifier is different than the root user identifier is. A password for the user identifier is provided to the third party to permit the third party to solve the problem with the computing system using the root privileges, via the user identifier temporarily having the root privileges to the computing system. The code patch is computer code installable on the computing system.

Temporarily providing higher privileges for computing system to user identifier

In a method for directing a computer network request to process through one or more components, a processor identifies a plurality of input/output (I/O) paths capable of serving a computer network request and a plurality of components along each path of the plurality of I/O paths. A processor predicts resource demands at each respective component of the plurality of components along each path of the plurality of I/O paths. A processor estimates expected delay times at each respective component of the plurality of components along each path of the plurality of I/O paths using predicted resource demands at each component. A processor determines an expected delay time for each path of the plurality of I/O paths based on the expected delay times at each respective component of the plurality of components along each path of the plurality of I/O paths.

Path selection for network service requests

A technique for implementing policy-aware backup and restore capability in a tiered storage system. If a data set's contents are backed up from the tiered storage system to a backup storage system, metadata for the data set may also be backed up. Prior to the data set being restored from the backup storage system to the tiered storage system, the backed up metadata is restored and processed to determine a tier among the tiered storage pools to which the data set will be restored.

Restoring a file to its proper storage tier in an information lifecycle management environment

A method and system for conducting an integrity check of system and user metadata when the metadata is mounted. A filesystem reviewer is embedded within software of a server node. A mutual exclusion lock is obtained on the selected metadata. An automated verification of the locked metadata is conducted while the filesystem is mounted. In addition, an automated salvage of the locked metadata may be conducted while the filesystem is mounted. The verification and/or salvage of the locked metadata can be applied to system, user, and/or a portion of the user metadata.

Mounted filesystem integrity checking and salvage

An apparatus, system, and method are disclosed for managing storage space allocation. The apparatus includes a recognizing module, a reserving module, and a managing module. The recognizing module recognizes a trigger event at a client of the data storage system. The reserving module reserves logical units of space for data storage. The management module manages the logical units of space at the client. Such an arrangement provides for distributed management of storage space allocation within a storage area network (SAN). Facilitating client management of the logical units of space in this manner may reduce the number of required metadata transactions between the client and a metadata server and may increase performance of the SAN file system. Reducing metadata transactions effectively lowers network overhead, while increasing data throughput.

Apparatus, system, and method for managing storage space allocation

Transitioning a data file in an archive system from read-write to write-once-read-many ('WORM') is triggered by a control file rather than using a timeout approach or a overload approach. A storage manager may receive a series of archive instructions to write a data file to an archive. During this time, the archive file may be in a read-write state. The storage manager may receive a control file, which may be structured, such as an XML file. Receipt of the control file signals to an storage manager archive system that the data file has been written to the archive and triggers the storage manager to make the data file immutable by switching it from read-write state to WORM state.

Transitioning an archived file to write-once-read-many

The acquisition of a lock among nodes of a divided cluster is disclosed. A method is performable by each of at least one node of the cluster. A node waits for a delay corresponding to its identifier. The node asserts intent to acquire the lock by writing its identifier to X and Y variables where another node has failed to acquire the lock. The node waits for another node to acquire the lock where the other node has written to X, and proceeds where Y remains equal to its own identifier. The node waits for another node to acquire the lock where the other node has written to a Z variable, and writes its own identifier to Z and proceeds where the other node has failed. The node writes a value to Y indicating that it is acquiring the lock, and maintains acquisition by periodically writing to Z.

Lock acquisition among nodes of divided cluster

Facilitating error handling of computing environments, including those environments having file systems. Responsive to an entity of the computing environment, such as a client of a file system, obtaining at least an indication of an error, a portion of functionality of the entity is automatically frozen. The obtaining is, for instance, responsive to an event of another entity of the computing environment, such as a server of the file system. Eventually, the frozen functionality is thawed allowing the functionality to proceed.

Automatically freezing functionality of a computing entity responsive to an error

A system, method and computer program product for implementing a distributed file system architecture requiring no direct access to user data from a metadata manager in a data storage network. Metadata operations requiring user volume access are performed via a proxy on a storage manager server that is a distributed file system client. Management of metadata can be isolated from user data because the metadata manager requires no capability to see the storage for user volumes. This allows for a distributed file management system to support heterogeneous environments without requiring any single point in the system to see all the storage.

File system architecture requiring no direct access to user data from a metadata manager

Techniques are provided for processing an Input/Output (I/O) request. At least one data block is allocated for use in completing the I/O request. A current operations record is stored for the I/O request. It is determined whether the I/O request has been completed within a specified period of time. In response to determining that the I/O request has not been completed within the specified period of time, the allocated at least one data block is fenced.

System and article of manufacture for fencing of resources allocated to non-cooperative client computers

Provided are a system and program maintaining information on a namespace comprised of filesets shared by clients over a network. Zone information is maintained on at least one zone, wherein each zone associates at least one client system, at least one fileset, and at least one storage pool. For one zone, the associated at least one fileset and at least one storage pool are accessible to the at least one client system. Clients are provided information on filesets included in a namespace, wherein each of a plurality of clients receive information on the at least one fileset associated with the client in the at least one zone including the client.

System and program for maintaining a namespace of filesets accessible to clients over a network

Growing community of inventors

Gresham, OR, United States of America

Thomas K Clark