Storage Technology Corporation

Oracle America, Inc.

International Business Machines Corporation

Oracle International Corporation

John Timothy O'Brien

George Franklin DeTar, Jr

Michael Steven Milillo

Jay S Belsan

Charles Allen Milligan

George A Rudeseal

Carl Thomas Madison, Jr

Kevin Lee Kidney

Scott E Burton

Kent D Prosch

David Glenn Hostetter

Craig A Buhlman

Alan R Permut

Richard Curtis Cleavinger

John Houkal

Neil L Thomas

Tracy D Dyer

Mitchell E Burghart

Hendrikus Everhardus Vandenbergh

Yinglei Wang

Nelson W Gildenmeister

Keith Allen Radebaugh

One or more embodiments include scheduling a new operation to be periodically executed. A start time for a new operation, to be periodically executed, is determined based on the start times of previously scheduled operations. The difference in value between a candidate start time and a start time of a previously scheduled operation is used to compute a cost for the candidate start time. A candidate start time, of all the candidate start times, with a lowest cost may be selected as the start time for the new operation.

Selection of a start time for a periodic operation

A data storage system includes a plurality of servers, a plurality of external interface providers each deployed on one of the servers, and a plurality of data storage controllers each deployed on one of the servers. Each of the providers is configured to receive a data access request, to identify one of the controllers that can satisfy the request, and to forward the request to the identified controller. The system further includes a plurality of data storage modules independent of the servers. The providers, controllers and modules are in communication with each other. Each of the controllers exclusively manages a portion of data content in at least one of the modules, and satisfies data access requests received from the providers by accessing the portion of data content in the at least one module.

Data storage system and method of processing a data access request

A data storage system manager includes one or more servers, at least one data collector deployed on at least one of the servers, at least one policy engine deployed on at least one of the servers, and at least one configuration manager deployed on at least one the servers. The at least one data collector is configured to collect resource utilization information including data storage wear rate of data storage system data storage modules. The at least one policy engine is configured to evaluate the collected information and to initiate changes to a configuration of the data storage system based on data storage wear rate and work load distribution policies. The at least one configuration manager is configured to implement the changes initiated by the at least one policy engine to control the data storage wear rate and a skew of the work load distribution within the data storage system.

Data storage system manager and method for managing a data storage system

A data storage system includes an input/output server and a storage unit. The input/output server includes a processor, memory, and a host channel adapter. The storage unit includes a processor, memory, and a storage module. The storage module includes a storage controller, and an interface block for connecting the storage module to a corresponding memory-mapped interface. The storage unit further includes a host channel adaptor. The storage unit host channel adapter is connected to a corresponding memory-mapped interface. The storage unit host channel adapter is capable of remote direct memory access to the input/output server. Protocol translation logic is configured to intercept a memory access request from the storage controller, and initiate a corresponding remote direct memory access to the input/output server through the storage unit host channel adapter and the input/output server host channel adapter.

Protocol translation in a data storage system

In a disk-based data storage system, a controller configured to control a reference count regeneration operation, the controller includes a control register, an address register, a status register, a boundary register, and an embedded memory. The control register may be configured to set up and initiate program instructions that are executed by at least one processor. The address register may be configured as a cache address pointer and correspond to at least one of a sort output list pointer, a virtual track table input list pointer, a reference list pointer, a track number table pointer, and a reference count mis-compare list pointer. The status register may be configured to indicate status of a routine. The routine includes at least one of a radix sort, a reference list count, a combine counts, and a merger of the reference list count into the TNT to generate an updated TNT. The boundary register corresponds to a disk cache memory that may be configured to store at least one of a reference list, a reference list count length, and a mis-compare list. The embedded memory may be configured to perform at least one buffer operation in connection with the routine.

System and method for reference count regeneration

In a data storage system in which there can be multiple references to a single instance of an object, a method for regenerating the number of references to each object instance. The method includes radix sorting the references to the objects to generate a reference list, counting the references to each unique object and merging the counts with the object descriptions, placing the count of the number of references to each object into the respective object description. The sorting, counting and merging techniques used by this method generate sequential memory access patterns that enable efficient use of low-cost memory and block-oriented memory access interconnect fabric protocols. Furthermore, multiple instances of the sorting, counting and merging processes can be used in parallel to reduce the time required to regenerate the reference counts for a large number of objects.

A method for configuring a computer network that includes a full duplex bi-directional first port and an initiator that can issue a request for the first port to assume a state. The first port, when in a first state, is able to bi-directionally communicate with a full duplex bi-directional second port, and when in a second state, is coupled to itself by having an output thereof coupled to an input thereof. The method comprises inhibiting the initiator from issuing the request, sending data to the initiator describing a desired state of the first port, and enabling the initiator to issue the request for the first port to assume the desired state.

Method and system for configuring a computer network

A method for detecting and remembering multiple sequential access patterns made from a host to a memory system having one or more logical storage devices. Once a sequential access pattern is detected, one or more tracks are requested to be prestaged ahead of the current access request. The number of tracks requested to be prestaged ahead may be responsive to the amount of storage available in the cache memory. One list is provided for each logical storage device. Each list has multiple entries and is logically divided into two parts. A sequential part contains entries for access streams which the system has determined are sequential and for which the system may have prestaged one or more tracks into cache. A candidate part contains information about recent host accesses which have not as yet been determined to be sequential. The entries within each of the two parts of the list are logically ordered in a most-recently-used fashion. The division between the two parts of the list is allowed to change so that although the total number of entries may be held constant, the proportion of the entire list allocated to each of the two parts can be varied. Inactive entries in the list are invalidated over time.

Sequential detection and prestaging methods for a disk storage subsystem

A system and method for providing a trusted machine state in a data processing system is disclosed. The data processing system includes a memory. The trusted machine state is stored in a first portion of the memory. The method and system include saving the trusted machine state in a second portion of the memory and reinitializing a portion of the memory. The portion of memory is separate from the second of memory. The method and system further include restoring the trusted machine state in the memory. According to the method and system disclosed herein, the data processing system may recover from failures and resume operation. Moreover, a non-disruptive code load, in which a new program can be loaded without disrupting system operations, may be performed.

Method and system for providing a trusted machine state

The present data file storage management system for snapshot copy operations maintains a two level mapping table which enables the data files to be copied using the snapshot copy process and only having to update a single corresponding mapping table entry when the physical location of the data file is changed. The snapshot copy updates to the contents of the first level of the two level mapping table are stored on the backend data storage devices to provide a record of the snapshot copy operation which can be used to recover the correct contents of the mapping table. This record of the snapshot copy operations remains valid even though the physical location of a copied data file instance is subsequently changed. Furthermore, the physical storage space holding the updated portions of the first level of the two level mapping table can be managed using techniques like those used to manage the physical storage space holding data file instances. Mapping table updates resulting from the snapshot copy operation are delayed until all mapping table updates resulting from earlier data file write operations have been completed and any attempt to update the mapping table to reflect data written to the original data file or the copy data file that occurs after initiation of the copy must wait until the first set of mapping table pointers have been copied.

Data file storage management system for snapshot copy operations

The disk drive array data storage subsystem maps between virtual and physical data storage devices and schedules the writing of data to these devices. The data storage subsystem functions as a conventional large form factor disk drive memory, using an array of redundancy groups, each containing N+M disk drives. This system copies data records by simply creating a duplicate data record pointer in a virtual track directory to reference the original data record. This enables the host processor to access the data record via two virtual addresses while only a single physical copy of the data record resides in the data storage subsystem.

Data record copy system for a disk drive array data storage subsystem

The present invention comprises a method for enhancing data integrity during write operations in a data storage subsystem which comprises an array of redundant, interconnected disk drives. A data storage subsystem is disclosed which provides for writing data segments spread across N disk drives of the array and provides for generating and recording redundancy information spread across M disk drives. Redundancy information is used for reconstructing erroneous data in the case of a disk drive failure. It may be desirable to write a final portion of the data last, after other portions of the data and associated redundancy information have been recorded to the disk drives. The last portion of the data segments may contain information that validates the data. It is common to automatically generate the redundancy information as the data is written to the disk drives. The generated redundancy information is then later written to the disk drives. This presents a problem in that the last portion of the data is written to disks before correct redundancy information is written. The present invention discloses a method of writing data and associated redundancy information to eliminate this problem. A first portion of the data excluding the validation portion is written to disks as the redundancy information is accumulated. Next, the last portion is applied to the redundancy accumulator circuitry but is not yet written to the disks. The redundancy information of the data is now recorded to the disks. Finally, the last portion of the data is recorded to the disks. This method eliminates the period of time in which apparently valid data segments on disks are actually invalid because the associated redundancy information is not yet written to disk.

Method for writing specific values last into data storage groups

The data record copy apparatus for a virtual memory system instantaneously makes a copy of a selected data record by simply generating a new set of pointers to reference the same physical memory location as the original reference pointer. All the pointers referencing the same physical memory space are stored in the virtual memory mapping tables to record all the virtual addresses of the data record. When the original data record is staged to an associated cache memory or a copy of the data record is in cache memory, the consistency of the data record is maintained by loading all the virtual addresses of the data record from the mapping tables to the hash table and collision list of the cache memory. Thus, any access of one of these virtual addresses will result in the virtual memory system locating the single data record instance in cache memory. If an associated host processor attempts to update the data record at one of these virtual addresses, the virtual memory system copies the data record to a new cache memory location so that a unique instance of the data record is available for update while the remaining virtual addresses still point to the original data record in cache memory.

Method and apparatus to access data records in a cache memory by

The use of a dynamically mapped virtual memory system permits the storage of data so that each data record occupies only the physical space required for the data. Furthermore, the data storage subsystem manages the allocation of physical space on the disk drives and does not rely on the file extent defined in the count key data format. Data compaction apparatus is provided to remove the gaps contained in the stream of count key data records received from the host processor. A data compression algorithm is then used to compress the received data into a compressed format for storage on the disk drives. It is the compacted, compressed data that is finally stored on the disk drives. Furthermore, any data record received from the host processor absent data in the user data field therein is simply listed in the virtual memory map as a null field occupying no physical space on the disk drives. The data storage control, through its mapping capability, stores the actual data in the minimum physical space required by overcoming the limitations imposed on large form factor disk drives by the use of count key data format data records. However, the data storage subsystem returns this stored data to the host processor in count key data format through a data record reformatting process once the stored compacted compressed data is staged to the cache memory for access by the host processor. The data storage subsystem is operationally independent of the host processor, yet performs as if it were a standard operationally dependent large form factor disk subsystem.

Apparatus for compressing data in a dynamically mapped virtual data

The early start mode data transfer apparatus coordinates the reading and writing of data files into and out of the data buffer of the tape control unit so that a data file can be written into the data buffer while another data file is concurrently being read out of the data buffer. Thus, both the host computer and the associated tape drive units can be active at the same time reading or writing data files into the data buffer of the tape control unit. The early start mode data transfer apparatus closely coordinates the reading and writing activity of the host computers and the associated tape drive units to avoid the overwriting data files in the data buffer as well as emptying the data buffer before the complete data file is written therein. By enabling the concurrent activity of the host computers and the associated tape drive units, the early start mode data transfer apparatus improves the efficiency of the data transfers between the host computers of the data transfers between the host computers and the associated tape drive units by increasing the occupany on the associated tape drive units. Thus, the associated tape drive units can provide a greater throughput since less time is spent in a wait mode while a data file is being written into or read out of the data buffer in the tape control unit. These and other advantages of this apparatus are explained.

Growing community of inventors

Louisville, CO, United States of America

John Timothy O'Brien