International Business Machines Corporation

Netapp, Inc.

Linda Ann Riedle

Sandeep Nirmale

Deepan Natesan Seeralan

Gregory Brian Pruett

Shah Mohammad Rezaul Islam

Chaitanya V Patel

David Benson Rhoades

Mayank Saxena

Kyle Diggs Sterling

Loellyn Jane Cassell

Laurent Nicolas Lambert

Brandon Taylor Long

Vikram Harakere Krishnamurthy

Daniel Curtis Wilson

William Joseph Piazza

Yong Cho

Philip Anthony Richardson

Bhoovaraghan Subramanian

Fani Jenkins

Manali Kulkarni

Dennis R Goetz

Bulent Abali

Simon C Chu

Dean Alan Kalman

Mohammad Banikazemi

Gregory Joseph McKnight

Dan Edward Poff

Gary Valentin

D Scott Guthridge

James W Arendt

Joaquin F Pacheco

Albert Andre Asselin

William W Buchanan, Jr

Richard Christopher Fore

Ziv Rafalovich

Charles Thorpe Stephan

Manali Kulkami

Abhishek Jain

Nancy M Wei

One or more techniques and/or systems are provided for mirror vote synchronization. For example, a first storage device is located at a first storage site, and a second storage device is located at a second storage site. The second storage device is configured according to a data mirroring configuration where data from the first storage device is mirrored to the second storage device. Mirror vote metadata is generated based upon an up-to-date state of the data mirroring configuration. The mirror vote metadata indicates whether the first storage device and/or the second storage device are up-to-date or not. The mirror vote metadata may be replicated between the first storage site and the second storage site. If the first storage site fails, then the second storage site may provide switchover operation using the second storage device based upon the mirror vote metadata.

Mirror vote synchronization

One or more techniques and/or systems are provided for controlling restoration of a storage aggregate. For example, a local storage device, located at a first storage site, and a remote storage device, located at a second storage site, may be assigned to a first storage aggregate. Responsive to a disaster of the first storage site, a gate may be created for the local storage device. The gate may block automated reconstruction and/or automated synchronization that may otherwise occur with respect to the local storage device. Until the local storage device is restored, the remote storage device may be used to service I/O requests that were otherwise directed to the local storage device. Responsive to receiving a user restoration command, the gate may be removed from the local storage device. Synchronization between the local storage device and the remote storage device may then be facilitated.

Storage aggregate restoration

One or more techniques and/or systems are provided for storage device health status synchronization. For example, a first health status registry is maintained for a first storage controller at a first storage site. The first storage controller stores storage device health status information of storage devices currently owned by or assigned to the first storage controller within the first health status registry. A second health status registry is maintained for a second storage controller at a second storage site. The second storage controller stores storage device health status information of storage device currently owned by or assigned to the second storage controller within the second health status registry. The first health status registry and the second health status registry are synchronized during operation of the first storage site and the second storage site and/or are merged between the first storage site and the second storage site after disaster recovery.

Storage device health status synchronization

In one aspect of the disclosure, progress identifiers (e.g., a plurality of flags) and/or generation numbers are utilized to determine a 'correct' non-volatile random access memory (NVRAM) replay source to use during a retried switchover or early switchback. Initially, a switchover sequence is initiated. The switchover sequence includes a replay of logged information to storage devices of an aggregate to ensure data consistency of the aggregate. Progress of the switchover sequence is tracked using the progress identifiers and/or generation numbers. After an interruption during the switchover sequence, switchover is retried or an early switchback is performed. The retried switchover or early switchback decides whether to replay logged information, and, if so, determines an NVRAM replay source to replay logged information from based on states of the progress identifiers and/or comparison of the generation numbers.

Providing data integrity in a non-reliable storage behavior

One or more techniques and/or systems are provided for load balancing between storage controllers. For example, a first storage controller and a second storage controller may be configured at a first storage site according to a high availability configuration, and may be configured as disaster recovery partners for a third storage controller and a fourth storage controller at a second storage site. If the first storage controller fails, the second storage controller provides failover operation for a first storage device. If a disaster occurs at the second storage site, the second storage controller provides switchover operation for a third storage device and a fourth storage device. Responsive to the first storage controller being restored, the third storage device may be reassigned from the second storage controller to the first storage controller for load balancing at the first storage site during disaster recovery of the second storage site.

Survival site load balancing

A system and method for automatically selecting and configuring storage devices according to an abstract partition table definition as part of a scripted operating system installation. The system has a deployment server in data communication with the one or more computers via the communication network. The deployment server has a deployment server storage device and a central processing unit. The deployment server storage device stores an inventory of the computers to which an operating system is to be deployed and a set of partition table requirements defined by one or more qualitative attributes associated with one or more computer storage device partitions. The central processing unit is in communication with the deployment server storage device and executes functions to map the inventory with the set of partition table requirements to create a partition table for each computer to which the operating system is to be deployed.

Method and system for creation of operating system partition table

Two methods (native and clone) are used for installing software, such as an operating system, on client system(s) booting from shared storage. The native installation method configures an interconnection network to create an exclusive communication zone between the client system and the shared storage system and installs the operating system on the client system using the exclusive communication zone. After the software is installed, the method terminates the exclusive communication zone. The clone installation method utilizes a point-in-time copy feature of the shared storage system to clone an operating system drive instantaneously. After the drive is cloned, it is logically attached to a new client and the operating system is customized for that client.

Method for installing operating system on remote storage: flash deploy and install zone

Growing community of inventors

Cary, NC, United States of America

Linda Ann Riedle