Emc IP Holding Company LLC

Emc Corporation

Douglas Eric LeCrone

Brett A Quinn

Paul A Linstead

Gary H Cox

Bruce Alan Pocock

Mark J Halstead

Benjamin W Yoder

Denis J Burt

Hana Moreshet

Daniel A Murphy

Eugene D Pflueger

David Meiri

Yuval Ofek

Natan Vishlitzky

Jeffrey L Jones

Kevin C Heasley

Ronald G Haupert

Marik Marshak

Vadim Longinov

Mike Specht

Martin Feeney

William R Fairchild

Bruce A Pocock

Haim Kopylovitz

Evan G Jones

Denis Hall

Elcana Cohen

William P Thibodeau

Michael J Scharland

Steven T McClure

Philip E Tamer

Michael E Specht

Ishay Kedem

Jerome Cartmell

Martin Farley

Izhar Sharon

Richard Goodwill

Shai Hess

Daniel P Murphy

Elizabeth E Patapoutian

Arieh Don

Adnan Sahin

Ian Wigmore

Dan Arnon

Dan Aharoni

Moshe Yanai

Ning Wu

Larry W McCloskey

Jeremy O'Hare

Ofer Michael

Daryl Francis Kinney

Yeshayahu Hass

Roger A Ouellete

Martin J Feeney

Roger A Ouellette

Isbay Kedem

Amir Holtzman

Raymond Wong

Michael John Egan

Stewart C Woodworth, Iii

Jeff L Jones

Larry W McLoskey

A host performs benchmark I/O operations for a storage system by an application on the host formulating a performance test command, providing the performance test command to a channel subsystem layer of the host that actuates communication hardware that provides signals to the storage system, and performing a plurality of benchmark I/O operations at the channel subsystem layer in response to a single performance test command by actuating the communication hardware that provides signals to the storage system. The single performance test command may result in actuating the communication hardware to read and write data at a plurality of addresses of the storage system. The channel subsystem layer may perform the plurality of benchmark I/O operations without receiving any additional data or commands from any applications on the host. A dispatch program that runs on the channel subsystem layer may receive operational data and may actuate the communication hardware.

Performance scale testing with I/O reuse

Backing up logical data from a storage system to a backup system includes accessing data on the storage system corresponding to logical file structures of data stored on the storage system, using machine learning to ascertain file access patterns and frequency of file backups, determining backup data based on the logical file structure of data stored on the storage system and the file access patterns and frequency of file backups, and directly transferring data corresponding the backup data from the storage system to the backup system. The logical file structures may be provided by file metadata. The file metadata may include at least one of: a VTOC (volume table of contents), an IXVTOC (indexed VTOC), VSAM (virtual storage access method) information, a VVDS (VSAM volume data set), file attributes, and/or catalog data. The file attributes may include management class information for each of the files.

Logical backup using a storage system

Transforming data that is provided by a first instance of an application that uses application-instance specific data includes determining if a component of the data is an application-instance specific component and, if the component is an application-instance specific component, transforming the component either at a storage system containing the data or as the component is being accessed by a second instance of the application, different from the first instance. Transforming the component at a storage system containing the data may be performed independently of any accesses of the data. Transforming the component at a storage system containing the data may be performed by the storage system. The first instance of the application may run on a first host and the second instance of the application may run on a second host different from the first host. The first and second instances of the application may run on a same host.

Transforming application-instance specific data

Copying data from a source storage system to a target storage system includes resetting a write tracker on the source storage system to track writes to the source storage system by one or more host computing systems, copying data from the source storage system to the target storage system after resetting the write tracker, suspending writes to the source storage system after copying the data, and copying data portions of the source storage system to the target storage system that are indicated as being written by the write tracker after suspending writes to the source storage system. Applications that write data to the source storage system may be quiesced in connection with suspending writes to the source storage system. Data portions may be repeatedly copied from the source storage system to the target storage system until an end condition is reached.

Copying data between storage systems

Handling I/O operations between a storage system and a host includes initiating a direct data transfer for each of the I/O operations that initially excludes other processes from using a CPU of the host, setting a first timer for each of the direct data transfers, converting at least some of the direct transfers to semi-synchronous I/O operations that release the CPU for use by other processes and transfer data directly between the storage system and the host in response to the first timer expiring prior to completion of a corresponding one of the direct data transfers, and setting a second timer that corresponds to an expected completion of the semi-synchronous I/O operation. The direct data transfers may exchange data between the host and cache memory of the storage system. The direct data transfers may be performed using a high speed connection between the storage system and the host.

Synchronous to semi-synchronous out-of-band i/o

Running an independent task in connection with a synchronous I/O operation between a storage system and a host includes starting the synchronous I/O operation, setting a timer for the synchronous I/O operation, starting the independent task that runs while waiting for completion of the synchronous I/O operation, and aborting the synchronous I/O operation in response to the timer expiring prior to completion of the synchronous I/O operation. The independent task may be ended in response to the timer expiring. The independent task may be ended in response to the I/O operation completing. The synchronous I/O operation may be performed using a high speed connection between the storage system and the host, which may be coupled to a smart network interface controller provided on a director board in the storage system. The smart network interface controller may include a system on a chip having a processor, memory, and non-volatile storage.

Running an independent task in connection with synchronous I/O operations

Creating point-in-time versions of files for applications at a storage system includes maintaining at the storage system a data structure correlating each of a plurality of applications with files for each of the applications and correlating each of the files with portions of the data storage system, using the data structure to determine specific files for a particular one of the applications for which a point-in-time version is being created, using the data structure to determine specific portions of the storage system corresponding to the specific files, suspending writes to the specific portions, completing previous writes to the specific portions following suspending writes, and performing a snapshot of the specific files following completing previous writes. The portions of the data storage may be extents. A host may provide information to the storage system to correlate applications with files and to correlate files with portions of the data storage system.

File level incremental continuous data protection

Obtaining a consistent set of snapshots of a group of storage volumes includes obtaining a plurality of snapshots, each of the snapshots being for one of the volumes, determining if there are any specific write operations to at least one volume that occurred after obtaining a first one of the snapshots and before obtaining a last of the snapshots, if there are any specific write operations, discarding the plurality of snapshots, and, if there are no specific write operations, designating the plurality of snapshots as the consistent set of snapshots. Obtaining a consistent set of snapshots may include repeatedly obtaining a plurality of snapshots and determining if there are specific write operations until either a particular condition is met or there are no specific write operations determined. The particular condition may be exceeding a predetermined number of iterations or a predetermined time limit.

Snapshots with selective suspending of writes

A host computing system includes an applications layer containing one or more user applications that perform I/O operations, an access methods layer that communicates with the applications layer, an I/O drivers layer that communicates with the access methods layer, and an SSCH simulation layer that communicates with the I/O drivers layer and that simulates a Fibre Channel connection that is accessed by applications in the applications layer. The host computing system may also include a TCP/IP stack layer that communicates with the SSCH simulation layer to provide TCP/IP communication for the host computing system. TCP/IP communication provided by the TCP/IP stack layer may be separate from any dedicated TCP/IP communication provided by the host. The host computing system may be coupled to a TCP/IP network. A cloud storage may be coupled to the network to communicate with the host computing system.

Simulation for alternative communication

Providing TCP/IP control commands to a storage system includes intercepting TCP/IP data provided on a TCP/IP network, determining if the TCP/IP data corresponds to a control command directed to the storage system, and converting the TCP/IP data to a Fibre Channel control command that is forwarded to the storage system if the TCP/IP data is a control command directed to the storage system. A host may intercept the TCP/IP data. The host may be coupled to the storage system. The storage system may have only Fibre Channel connectivity. A PC having only TCP/IP connectivity may provide the control commands via the TCP/IP network. At least two hosts may be coupled to the TCP/IP network. Software on one of the hosts may provide the control commands for the storage system. The storage system may be coupled to an other one of the hosts.

Singular control path for mainframe storage

Using high speed data transfer protocol to transfer data between a host and a storage system includes determining if a portion of data is a candidate for being transferred using the high speed data transfer protocol if the portion is not a candidate for high speed data transfer, transferring the data using a relatively low speed data transfer protocol if the portion is a candidate for high speed data transfer, attempting to transfer the data using the high speed data transfer protocol, and if transferring the data using the high speed data transfer protocol is unsuccessful, transferring the data using the relatively low speed data transfer protocol. The high speed data transfer protocol may use a first connection and the relatively low speed data transfer protocol may use a second connection. The first connection may be a PCIe or InfiniBand connection. The second connection may be a FICON connection.

Using high speed data transfer protocol

Using an alternative communication protocol between a first system and a second system that are otherwise configured to communicate using a FICON protocol includes the first system determining if the alternative communication protocol is handled by the second system, the first system providing encapsulated data by encapsulating FICON data if the alternative communication protocol is handled at the second system, and the first system transmitting the encapsulated data directly to the second system using the alternative communication protocol if the alternative communication protocol is handled at the second system. The alternative communication protocol may be TCP/IP. At least one of the systems is a host computing system, an array storage system, and/or a tape emulation system. At least one of the systems may be a simulation of a host computing system, an array storage system, and/or a tape emulation system.

Encapsulated FICON communication

A backup data storage system includes non-volatile memory units, a disk interface coupled to at least some of the non-volatile memory units, a connection component that facilitates exchanging data with the backup data storage system, and a smart network interface controller, coupled to the disk interface and the connection component to provide tape emulation to a host coupled to the backup data storage system. The disk interface, the connection component, and the smart network interface controller may be coupled using a PCIe bus. Tape data written to the backup storage device may be stored on the non-volatile memory units. A processor coupled to the smart network interface controller and the disk interface may receive the data from the smart network interface controller and may provide the data to the disk interface to store the data on the non-volatile memory units. The connection component may be a FICON connection component.

Using a smart network interface controller with storage systems

Accessing additional storage space of a storage system includes reading a physical data fragment that is an incremental subset of a physical storage unit of the storage system, accessing metadata corresponding to logical blocks stored on the physical data fragment to determine a sidebar storage portion of the physical data fragment that is unused by the logical blocks, and accessing data of the sidebar storage portion. Accessing data of the sidebar storage portion may include providing data from the sidebar storage portion to a calling process. Accessing data of the sidebar storage portion may include modifying a portion of data from the physical data fragment and writing the physical data fragment back to the physical storage device. The physical data fragment may be locked prior to reading the physical data fragment and the physical data fragment may be unlocked after writing the physical data fragment.

Accessing storage space of a physical data fragment in a storage system

Providing continuous data protection includes maintaining a database having substantially all data modifications made to a primary volume over a recovery interval. The database is maintained in conjunction with a copying operation where the data of the primary volume are mirrored to a remote volume to permit recovery of mirrored data in the event of loss of primary volume data. The contents of the remote volume generally lag behind the contents of the primary volume by substantially the recovery interval. Providing continuous data protection also includes providing data roll-back to a precise point in time within the recovery interval by applying, to the contents of the remote volume, all data modifications in the database that occurred between the latest data modification to the remote volume and the precise point in time within the recovery interval. A time stamp mechanism of sufficient precision and granularity may be used.

Continuous data protection

Growing community of inventors

Hopkinton, MA, United States of America

Douglas Eric LeCrone