Hewlett-Packard Development Company, L.p.

Digital Equipment Corporation

Other

Western Digital Technologies, Inc.

Compaq Computer Corporation, Inc.

Seagate Technology Incorporated

Hewlett-Packard Development, L.p.

Xiotech Corporation

Susan Gaye Elkington

Stephen J Sicola

Clark Edward Lubbers

Randy Hess

Anuja Korgaonkar

James Francis McCarty

Jason Leveille

Michael D Walker

Richard Franklin Lary

Michael Walker

Roger Oakey

Randall L Hess

James Edwin Pherson

Jesse L Yandell

Randy Roberson

Kenneth Hoffman Bates

Paul D Guttormson

Randal S Marks

Wayne H Umland

Julia A Hodges

Gerald L Golden

Ryan J Johnson

Michael Stillwell

Bert Martens

Karen E Workman

John Franklin Mertz

Bradford Scott Morgan

Theodore E Bruning, Iii

James S Jackson

James Perry Jackson

Ronald H McLean

Thomas F Fava

Wayne Umland

John Fandel

Jason Richard Huddleston

Enhancements for managing quality of service in a multi-host Peripheral Component Interconnect Express (PCIe) switching environment involve a host system configured to maintain quality of service statistics corresponding to data interactions with a PCIe storage device available via a PCIe switch. The host system may further receive secondary quality of service statistics for one or more other host systems communicatively coupled to the PCIe device via the PCIe switch, and determine a maximum queue depth for the host system based on the quality of service statistics and the second quality of service statistics to maintain a quality of service for at least the host system.

Queue depth management for host systems accessing a peripheral component interconnect express (PCIe) device via a PCIe switch

Described herein are enhancements for managing quality of service in a multi-host Peripheral Component Interconnect Express (PCIe) switching environment. In one implementation, a host system is configured to maintain quality of service statistics corresponding to data interactions with a PCIe storage device available via a PCIe switch. The host system may further receive secondary quality of service statistics for one or more other host systems communicatively coupled to the PCIe device via the PCIe switch, and determine a maximum queue depth for the host system based on the quality of service statistics and the second quality of service statistics to maintain a quality of service for the host systems.

An apparatus and associated method including a first storage device and a second storage device, each coupled to a remote server independently of the other via a network. Resilient mirroring logic is stored in each of the storage devices that establishes a peer-to-peer communication connection with the other storage device in response to receiving a data access command from the remote server.

Resilient mirroring

Resilient mirroring utilizing peer-to-peer storage

A method for communicating between devices in a network includes creating an I/O tunnel between a first device and a second device through the network. The I/O tunnel is associated with I/O resources in both the first device and the second device, and wherein at least one of the I/O resources comprises a buffer resource. A data transfer operation may be sent between the first device and the second device by consuming at least some of the I/O resources associated with the I/O tunnel. A plurality of commands or a plurality of responses can be aggregated into a single buffer resource. Upon completion of the data transfer operation, the I/O resources that are consumed are automatically renewed.

Data transfer protocol for data replication between multiple pairs of storage controllers on a SAN fabric

A system for communicating between two devices in a network in which a semi-persistent tunnel is established between the two devices in advance of data communication. The semi-persistent tunnel includes resources that are pre-allocated in a first device at a first end of the communication link by a second device at the second end of the communication link. The first and second devices implement a plurality of processes for handling data transfer operations. Preferably, the semi-persistent tunnel also includes resources that are pre-allocated in a device at the second end of the communication link by the device at the first end of the communication link to allow bi-directional communication. Data transfer operations transmitted through the tunnel include an identification of specific resources of the pre-allocated resources that are to handle the data transfer operation. Data transfer operations also include a vector identifying a particular process among the plurality of processes that implement the data transfer operation.

Data transfer protocol for data replication between multiple pairs of storage controllers on a san fabric

In one embodiment, a method comprises receiving, in a source controller, a signal indicative of a write request to a source volume managed by the source controller in response to the signal: writing data associated with the write request to a destination controller when a connection to the destination controller is available; and setting a synchronization flag associated with the data storage segment managed by the source controller when a connection to the destination controller is unavailable.

Data synchronization management

A method of managing critical resource usage comprises monitoring queue depth in at least one critical resource queue for a high level of resource consumption and preventing subsequent commands of a client/server combination in response to a command that increases resource consumption to the high level. The method further comprises queuing an identifier of the client/server combination on a waiting queue, removing the client/server combination identifier from the waiting queue in queuing order as resource consumption declines, and enabling commands of the client/server combination.

Resource management system

A data storage system adapted to maintain redundant data storage sets at a destination location(s) is disclosed. The data storage system establishes a copy set comprising a source volume and a destination volume. Data written to a source volume is automatically copied to the destination volume. The data storage system maintains a data log that may be activated when the destination volume is inaccessible due to, for example, a malfunction in the destination storage system or in the communication link between the source system and the destination system. I/O commands and the data associated with those commands are written to the data log, and after a destination system becomes available the information in the data log is merged into the destination volume to conform the data in the destination volume to the data in the source volume. The data log competes for disk capacity with other volumes on the system, and log memory is allocated as needed. In operation, the array controllers monitor connection status between array controllers in a DRM group. When a connection failure is detected the controllers determine which of the remaining controllers determine whether the source controller remains online, and if not then which destination controller should assume the role of the source controller.

Automatic site failover

A system in which data operations for a data replication group are received in-order, and buffered. When the operations are complete, they are stored in a non-volatile memory atomically with a group sequence number. The cache is preferably mirrored. This creates a persistent association between the data operation and the sequence number. After the atomic store is performed in at least one non-volatile cache, the data operation is propagated to another member of the data replication group along with the group sequence number. In the other member, the data operation is cached at least once in a non-volatile cache atomically with the group sequence number. In this manner, the set of group sequence numbers for a plurality of operations forms a continuous ordering of the data operations.

System and method for maintaining order for a replicated multi-unit I/O stream

A data storage system is provided with a pair of controllers and circuitry configured for failing back from a single active write back mode to a dual active write back mode by copying cached data directly from a cache of a survivor controller of the pair of controllers to a cache of the other controller. A method is provided for failing over from a dual active mode of first and second controllers to a single active mode of the first controller by relying on previously mirrored cache data by the second controller; reinitializing the second controller; and failing back to the dual active mode by copying cached data directly from the first controller to the second controller.

Failover and failback of write cache data in dual active controllers

A very large virtual volume (e.g., in excess of 500 GB) is formed by distributing the disks in eleven, six-disk RAID-5 sets across the six busses of a primary local back-end controller. A spare disk is provided on each of the six busses. Each RAID-5 set is protected from the failure of a single disk by the spare disks on the busses, which can use the parity data stored in a RAID-5 set to rebuild the data stored on a failing disk and thereby restore redundancy to the RAID-5 set. Each RAID-5 set is also protected from the failure of a bus by the parity inherent in RAID-5. The RAID-5 sets are striped by a front-end controller connected to the primary local back-end controller, and the striped RAID-5 sets are presented to a host computer as a very large virtual volume. If the individual disks are 9.1 GB in size, the size of the very large virtual volume can reach 500.5 GB. If desired, additional groups of eleven, six-disk RAID-5 sets can be formed on additional back-end controllers for purposes of redundancy, cloning (which generates a copy of the data that can be used for off-line backup without interrupting read/write activities on the virtual volume), and disaster tolerance through remote storage. These additional groups of RAID-5 sets along with the RAID-5 sets from the primary local back-end controller, can be formed into mirror sets by the front-end controller, which then stripes the mirror sets and presents the striped mirror sets to the host computer as the very large virtual volume.

Apparatus and method for providing very large virtual storage volumes using redundant arrays of disks

A data storage system adapted to maintain redundant data storage sets at a destination location(s) is disclosed. The data storage system establishes a copy set comprising a source volume and a destination volume. Data written to a source volume is automatically copied to the destination volume. The data storage system maintains a data log that may be activated when the destination volume is inaccessible due to, for example, a malfunction in the destination storage system or in the communication link between the source system and the destination system. I/O commands and the data associated with those commands are written to the data log, and after a destination system becomes available the information in the data log is merged into the destination volume to conform the data in the destination volume to the data in the source volume. The data log competes for disk capacity with other volumes on the system, and log memory is allocated as needed. If multiple remote sites are provided, then the system permits to implements a merge fence to manage the process of merging data into multiple destinations.

System and method for multi-destination merge in a storage area network

Methods and systems for managing disk capacity allocated to a data log in a source data storage system during a merge process are disclosed. Data in the data log may be merged into corresponding data on a destination storage system connected to the source data storage system by at least one communication link. In one embodiment a method comprises maintaining a ratio of merge writes out of the data log to writes from a host computer into the data log within a desired range until the write process reaches a predetermined distance from the end of the data log, and quiescing writes from a host computer into the data log until the data log is fully written to the destination storage system.

System and method for ensuring merge completion in a storage area network

A disaster-tolerant data backup and remote copy system which is implemented as a controller-based replication of one or more LUNs (logical units) between two remotely separated pairs of array controllers connected by redundant links. In the situation wherein an array controller fails during an asynchronous copy operation, the partner array controller uses a 'micro log' stored in mirrored cache memory to recover transactions, in order, which were 'missed' by the backup storage array when the array controller failure occurred.

Method for transaction log failover merging during asynchronous operations in a data storage network

Growing community of inventors

Colorado Springs, CO, United States of America

Susan Gaye Elkington